whr 2010 background paper 48 chi 2010 soors et al
DESCRIPTION
Background paperTRANSCRIPT
Community Health Insurance and Universal Coverage: Multiple paths, many rivers to cross Werner Soors, Narayanan Devadasan, Varatharajan Durairaj and Bart Criel
World Health Report (2010)Background Paper, 48
The path to universal coverageHEALTH SYSTEMS FINANCING
© World Health Organization, 2010 All rights reserved. The designations employed and the presentation of the material in this publication do not imply the expression of any opinion whatsoever on the part of the World Health Organization concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. Dotted lines on maps represent approximate border lines for which there may not yet be full agreement. The mention of specific companies or of certain manufacturers' products does not imply that they are endorsed or recommended by the World Health Organization in preference to others of a similar nature that are not mentioned. Errors and omissions excepted, the names of proprietary products are distinguished by initial capital letters. All reasonable precautions have been taken by the World Health Organization to verify the information contained in this publication. However, the published material is being distributed without warranty of any kind, either expressed or implied. The responsibility for the interpretation and use of the material lies with the reader. In no event shall the World Health Organization be liable for damages arising from its use. The findings, interpretations and conclusions expressed in this paper are entirely those of the author and should not be attributed in any manner whatsoever to the World Health Organization.
I
Community Health Insurance and Universal Coverage: Multiple paths, many rivers to cross
World Health Report (2010) Background Paper, No 48
Werner Soors1, Narayanan Devadasan2, Varatharajan Durairaj3 and Bart Criel1
1 Unit of Health Policy and Financing, Institute of Tropical Medicine, Antwerp, Belgium 2 Institute of Public Health, Bangalore, India 3 Department of Health Systems Financing, World Health Organization, Geneva, Switzerland
II
CONTENTS Framing and assessing Community Health Insurance 1
within universal coverage
The scope of Community Health Insurance 4 in low- and middle-income countries
Community Health Insurance in Africa 7
Community Health Insurance in West Africa 9
Senegal 11 Mali 13 Guinea 19 Burkina Faso 21 Benin 22 Togo 24 Cameroon 25 Niger 26 Mauritania 27 Ghana 29
Community Health Insurance in East and Central Africa 35
The United Republic of Tanzania 36 Kenya 38 Uganda 39 Rwanda 41 Burundi 46 The Democratic Republic of Congo 50
Community Health Insurance in Asia 53
China 54 India, Nepal and Bangladesh 63 Cambodia 73 The Lao People’s Democratic Republic 77 The Philippines 79 Indonesia 80
Community Health Insurance in Latin America 83
Promises and challenges of Community Health Insurance 84 at the crossroads of universal coverage
References 88
III
ABBREVIATIONS
ACCORD Action for Community Organisation, Rehabilitation and Development (development NGO, Gudalur, Tamil Nadu, India)
AMEL Assurance Maladie des Elèves (CHI scheme for school children, Mali)
AMO Assurance Maladie Obligatoire (mandatory health insurance, Mali)
AMV Assurance Maladie Volontaire (standard CHI package from UTM, Mali)
ANC Antenatal care ASaCo Association de Santé Communautaire
(community health board, Mali) Askeskin Asuransi Kesehatan Masyarakat Miskin
(Health Insurance for Poor Population, targeted through a health benefits card; Indonesia, 2004-2005 on)
ASHWINI Association for Health Welfare in the Nilgiris (NGO, hospital and CHI scheme under ACCORD,
Gudalur, Tamil Nadu, India, 1992 on)
BAIF Bharat Agro Industries Foundation (development and research NGO in India, 1967 on; with a CHI scheme in rural Pune, Maharashtra, 2001 on) BOF/FBS Belgisch Overlevingsfonds/le Fonds Belge de Survie (Belgian Survival Fund) BPL Below-poverty-line (referring to the poverty threshold used by the government of India)
BTC/CTB Belgische Technische Coöperatie/Coopération technique belge (development cooperation agency, Belgium) BMI Basic Medical Insurance (compulsory urban health insurance, also known as UEBMI,
[formal sector and government employees], China, 1998 on) CAAFW Cambodian Association for Assistance to Families & Widows (Cambodian NGO, 1998 on; with a HEF, 2000 on, and a CHI scheme, 2003 on) CAM Carte d’Assurance Maladie (health insurance card, CHI scheme, Burundi, 1984 on) CBHI Community-based health insurance (widely used synonym for CHI) CCS Civil Servants’ Scheme (social health insurance scheme, public formal sector, Lao PDR, 2006 on) CGHS Central Government Health Scheme (social health insurance scheme for civil servants, India, 1954 on)
CHE Catastrophic Health Expenditure CHF Community Health Fund (district-based CHI scheme, Tanzania, 1996 on) CHI Community Health Insurance CIC Co-operative Insurance Company (Kenya, operates CHI schemes, 2001 on)
CIDR Centre International de Développement et de Recherche (development and research agency, active in 12 African countries, France)
IV
CMDT Compagnie Malienne pour le Développement des Textiles (national cotton company, Mali) CMS Cooperative Medical Scheme (village CHI scheme, expression of RMCS, China)
CNC Cadre National de Concertation des acteurs du développement des mutuelles de santé (national forum of exchange between CHI actors, several countries) CONSAMUS Concertation Nationale des Structures d’Appui aux Mutuelles de Santé (national forum of exchange between CHI actors, Benin)
CSCom Centre de Santé Communautaire (community health centre, Mali)
Dana sakit Sickness Funds (experimental CHI scheme, Solo, Indonesia, 1963-1969) Dana sehat Health Funds (CHI schemes, Indonesia, 1969 on) DFID UK Department for International Development Dhan Development of Human Action Foundation (development NGO, Tamil Nadu, India, 1997 on) DPS Direction de la Protection Sociale et de l’économie solidaire (directorate of social protection, Mali)
DWMHI District Wide Mutual Health Insurance (district-based scheme under NHIS, Ghana, 2005 on) EICV Enquête Intégrale sur les Conditions de Vie des ménages au Rwanda (Integrated Living Conditions Survey, Rwanda;
EICV1 2000-2001, EICV2 2005-2006) ESIS Employees’ State Insurance Scheme (social health insurance scheme for formal-sector workers, India, 1948 on)
FAM Fonds d’Assistance Médicale (social assistance in health, Mali, 2004 on)
Fbu Franc burundais (Burundian franc, monetary unit)
GIS Government Insurance Scheme (social health insurance scheme for civil servants,
gradually integrated into the BMI from 1998 onwards, China) GRET Groupe de recherche et d’échanges technologiques (French NGO, implementing SKY in Cambodia) GTZ Gesellschaft für Technische Zusammenarbeit (development cooperation agency, Germany) HEF Health Equity Fund (social fund for free access to healthcare for the extreme poor) HmI Health micro-Insurance HMO Health Maintenance Organisation ILO International Labour Organization ILO-STEP Strategies and Tools against social Exclusion and Poverty
(programme of the International Labour Organization)
IRDA Insurance Regulatory and Development Authority (India) JPKM Jaminan Pemeliharaan Kesehatan Masyarakat (health maintenance organisations, Indonesia, 1992 on) JRHIS Jowar Rural Health Insurance Scheme (CHI scheme in Maharashtra, India, 1981 on) Kartu sehat Health benefits card for the poor
V
(Indonesia, 1998 on) Karuna Trust Development NGO in Karnataka, India (with CHI scheme, 2002 on)
KCBHFA Kenya Community Based Health Financing Association (platform for CHI actors, Kenya, 2002 on) KKPKP Kagad Kach Patra Kashatakari Panchayat (waste pickers collective in Pune, Maharashtra, India, 1993 on, with a CHI scheme, 2003) KKVS Kadamalaj Kalanjiam Vattara Sangam (micro-finance programme and CHI scheme of the Dhan Foundation,
Tamil Nadu, India, 2000 on) La Concertation La Concertation entre les acteurs du développement des mutuelles de santé en Afrique (CHI network, francophone Africa) LIS Labour Insurance Scheme (social health insurance scheme for state-owned enterprise workers, gradually integrated into the BMI from 1998 onwards, China) MSAG Mutuelles de Santé de l’Archidiocèse Gitega (CHI network, Burundi)
MCSDN Mutuelle Communautaire de Santé de Dar-Naïm (CHI scheme in Dar-Naïm, Nouakchott, Mauritania, 2003 on,
with a HEF added, 2005 on)
MFA Medical Financial Assistance (social assistance in health, 2003 on, urban & rural China) MHO Mutual Health Organisation (synonym for CHI scheme, analogue of the French mutuelle de santé)
MURIGA Mutuelle pour les Risques liés à la Grosesse et à l’Accouchement (CHI scheme for pregnant women, Guinea) MUSCABU Mutuelles de Santé des caféiculteurs du Burundi (CHI network, Burundi)
MUTCO Mutuelle Cotonnière de Nongon (rural CHI scheme, Mali)
Navsarjan Trust Development NGO in Gujarat (with CHI scheme, 1999 on)
NCEUS National Commission for Enterprises in the Unorganised Sector (originally called National Commission for Enterprises in the Unorganised/Informal Sector, India, 2004 on)
NCMS New Cooperative Medical System also known as new Rural Cooperative Medical System (new RCMS, NRCMS) (voluntary rural health insurance, China, 2003 on) NDC National Democratic Congress (political party, Ghana; governed 1993-2000 and 2009-today)
new RCMS (also NRCMS) new Rural Cooperative Medical System, also known as New Cooperative Medical System (NCMS) (voluntary rural health insurance, China, 2003 on) NGO Non-governmental organisation NHIA National Health Insurance Act (Ghana, 2003) NHIC National Health Insurance Council, ‘the Council’ (regulating body for health insurance, Ghana, following the 2003 NHIA) NHIF National Health Insurance Fund, ‘the Fund’
VI
(financial pool, Ghana, following the 2003 NHIA) National Health Insurance Fund (social health insurance scheme for civil servants, Tanzania, 2001 on)
NHIL National Health Insurance Levy (earmarked portion of VAT, Ghana, following the 2003 NHIA) NHIS National Health Insurance Scheme (Ghana, following the 2003 NHIA, started operations in 2005) (Uganda, still in design) NHS National Health Service NIC National Insurance Company (public insurer, India) Nidan Development NGO, social security programme and CHI scheme (Patna, Bihar, India; NGO 1996 on, CHI 1999 on) NISR National Institute of Statistics of Rwanda NPP New Patriotic Party (political party, Ghana; governed 2001-2008 and introduced the NHIS) NRHM National Rural Health Mission
(national programme to improve the availability of and access to quality healthcare by the rural poor, India, 2005-2012)
NSHIF National Social Health Insurance Fund (projected social health insurance scheme, enacted in 2004, Kenya) NSSF National Social Security Fund (Cambodia, 2008 on) NUHM National Urban Health Mission (national programme to improve the availability of and access to quality healthcare by the urban poor, India, announced in 2008) PARESOC Programme d’action régionale pour l’économie sociale (support initiative for the informal sector, in Benin, Burkina Faso and Mali) PBF Performance-based financing PhilHealth Philippine Health Insurance Corporation (national social health insurance agency, Philippines, 1995 on) PHR Partners for Health Reform (USAID project, until 2000)
PHRplus Partners for Health Reformplus (USAID project, since 2000, also known as Health Systems 20/20) Plan Sésame Government exemption scheme for the elderly
(Senegal)
PNDS Plan National de Développement Sanitaire (national health sector development plan, Burundo, 2006-2010) PNPMS Programme National de Promotion des Mutuelles de Santé (national programme for the promotion of community health insurance,
DR Congo) PPS Preferred Provider System PRIMA Projet de recherche sur le partage du Risque Maladie (CHI research project, Guinea)
PROMUSAF Programme d’appui aux Mutuelles de Santé en Afrique (CHI support initiative, in Benin, Burkina Faso, Mali and Senegal)
VII
PSDN Projet Santé Dar-Naïm (primary health project in Dar Naïm, Nouakchott, Mauritania, with a CHI scheme [MCSDN] and a HEF)
QUIBB Questionnaire des indicateurs de base du bien-être (Core Welfare Indicators Questionnaire, Burundi, 2006) RAHA Raigarh Ambikapur Health Association (NGO – 1969 on – and CHI scheme – 1980 on, Chhattisgarh, India)
RAMA La Rwandaise d’assurance maladie (social health insurance scheme for civil servants, Rwanda) RAMED Régime d’Assistance Médicale
(social health assistance for the extreme poor, Mali, 2010 on) RAMS Réseau d’appui aux mutuelles de santé de Burkina Faso (CHI support initiative, Burkina Faso) RAS Réseau Alliance Santé (CHI support initiative, Benin) RCMS Rural Cooperative Medical System (rural health services’ financing and organisational design, China) RMHC Rural Medical Health Care (experimental voluntary rural health insurance, China, 2003-2006) RSBY Rashtriya Swasthya Bima Yojana (national health insurance scheme for BPL people, India, 2008 on) SEWA Self-Employed Women’s Association (labour union, Gujarat and other states, India) SHH Student’s Health Home (NGO, hospital and CHI scheme in West Bengal, India, 1952 on)
SHI Social Health Insurance SHIELD Strategies for Health Insurance for Equity in Less Developed countries (research project in Ghana, Tanzania and South Africa, 2006-2010) SHINE Social Health Insurance Networking and Empowerment project (GTZ project, the Philippines, 1995-2001)
SKY Sokapheap Krousat Yeugn (Khmer for Health for Our Families) (CHI scheme, Cambodia, 1998 on) SSNIT Social Security and National Health Insurance Trust (social insurance for formal sector workers, Ghana)
SSO Social Security Organisation (social health insurance scheme, private formal sector, Lao PDR, 2001 on) TNCHF Tanzanian Network of Community Health Funds (CHI umbrella organisation, Tanzania, 2003 on)
UCBHFA Uganda Community Based Health Financing Association (CHI umbrella organisation, Uganda, 1998 on) UEBMI Urban Employee Basic Medical Insurance (mandatory urban health insurance
[formal sector and government employees], China, 1998 on) UEMOA Union Economique et Monétaire Ouest Africaine (West African Economic and Monetary Union)
UMU Uganda Martyrs University UNDP United Nations Development Programme UNICEF United Nations Children’s Fund
VIII
UpLift UpLift India Association (association of NGOs for social protection, grouped in UpLift Family, UpLift Health and UpLift Wealth, India, 2001 on)
UpLift Health CHI branch of UpLift (first scheme in urban Pune, Maharashtra, 2002 on) URBMI Urban Resident Basic Medical Insurance (voluntary urban health insurance
[for those not covered by (UE)BMI], China, 2007 on) USAID U.S. Agency for International Development UTM Union Technique de la Mutualité Malienne
(CHI umbrella organisation, Mali) VAT Value-added tax VHS Voluntary Health Services (NGO, hospital and CHI scheme in Tamil Nadu, India, 1972 on) WHA World Health Assembly Yeshasvini Yeshasvini Cooperative Farmers Health Scheme (CHI scheme, Karnataka, India, 2003 on) ZAER Zone d’Animation et d’Expression Locale
(local unit of the national cotton company CMDT, Mali)
IX
CREDITS & ACKNOWLEDGEMENTS
Werner Soors did the literature review, drafted and finished this paper. Narayanan
Devadasan and Varatharajan Durairaj gave inputs and commented in every stage of the
writing. Bart Criel gave inputs, commented and took care of overview and coordination. All
four authors contributed to and approved the final version of the text.
The authors express their gratitude for inputs and comments received from a number of
colleagues, among them Bart Jacobs, Bruno Marchal, Fahdi Dkhimi, Joris Michielsen,
Josefien van Olmen, Maria-Pia Waelkens, Pascal Ndiaye and Prashanth NS.
The authors thank WHO for the opportunity to produce this work, and for their patience.
This document is not a formal publication of the World Health Organization.
The body of knowledge synthesised in this paper is the result of the efforts of thousands in
this world. Omission and errors are the sole responsibility of the authors.
X
Framing and assessing Community Health Insurance within universal coverage
This paper examines community health insurance (CHI) as a means to an end, assessing CHI
in the itinerary towards universal coverage. In 2005, the World Health Assembly (WHA)
explicitly urged its member states to strive and plan for universal coverage, within the
particular macroeconomic, socio-cultural and political context of each countryi. The WHA
did so adopting its secretariat’s definition of universal coverage as “access to key promotive,
preventive, curative and rehabilitative health interventions for all at an affordable cost”ii.
The 2008 World Health Report defines universal coverage as “universal access to the full
range of personal and non-personal health services they need, with social health
protection”iii. Earlier, Nitayarumphong had defined universal coverage as “a situation where
the whole population of a country has access to good quality services according to needs and
preferences, regardless of income level, social status, or residency”iv, a definition later
embraced by the Commission on Social Determinants of Health that added the requirement
“that people are empowered to use these services”v.
While these definitions differ in their precision of the scope of services to be covered – what
Kutzin calls the relative concept of universal coverage with respect to healthcare servicesvi –
they are absolute with respect to population coverage. All three definitions refer to
healthcare coverage in line with the ultimate goal of achieving health for alli,iv, which is
broader than mere insurance coverage. Indeed, as Mills pointed out, “inclusion within a
financing scheme does not guarantee access to benefits”vii. In each of these definitions,
universal coverage evokes equity in access, through financial risk protection and implicitly
associated with equity in financing. We explicitly adhere to these concepts of universal
coverage as healthcare coverage when framing and assessing CHI, to avoid confusion with
more restrictive but incrementally used terms as universal insurance coverage (and/or ‘basic
1
universalism’ in Latin Americaviii).
We give a broad overview of the scope and origin of CHI in low- and middle-income
countries, focusing deliberately on six country cases we deem to be of singular interest:
Senegal, Mali, Ghana, Rwanda, China and India. We specifically consider three dimensions,
two determinants and one specific dynamic within the itinerary towards universal coverage,
applied to CHI.
The three dimensions we consider are the breadth, depth and height of coverage, as spelled
out in the 2008 World Health Reportiii.
Breadth (alternatively called width) is short for population coverage; depth for service
coverage, referring to the range of services covered; and height for financial coverage,
referring to the proportion of costs covered. We start the description on breadth, depth and
height of coverage in each country section with data from World Health Statistics 2010: the
share of prepayment plans (including CHI, not including mandatory health insurance) in
private health expenditure, and out-of-pocket expenditure as a proportion of private health
expenditure, in the years 2000 and 200735. These data are obviously neither specific to CHI
nor to healthcare coverage. We use them to frame CHI in a national picture of health
financing, not as an indication of success or failure of CHI. Often they are the only available
proxy for financial protection (or lack of it) through CHI at national level; never are they a
comprehensive indicator of healthcare coverage.
The two determinants of universal coverage we consider are institutional design and
organisational practice, as elaborated by Carrin, Mathauer and colleaguesix, ,x xi based on the
conceptual work of Northxii. In line with Mathauer and Carrin, we posit that the progress
towards universal coverage is contingent on interconnected rules and organisationsx.
Otherwise stated, quoting North, “if institutions are the rules of the game, organisations are
2
the players”xii.
The specific dynamic we consider is one of empowerment and transformation: a process of
building capabilities and claiming rights, ultimately leading to a social compact of equity
and solidarityxiii, , , , ,xiv xv xvi xvii xviii. In the case of CHI, part of this process is what Diop and Ba
recently termed ‘political expansion for social inclusion’xix.
Finally – based on our framework of three dimensions, two determinants and one dynamic –
we aim at lessons learnt and summarise the promises and challenges of CHI at the
crossroads of universal coverage.
3
The scope of Community Health Insurance in low- and middle-income countries
Community health insurance covers a wide variety of health insurance arrangements – with
vast gradients in terms of ownership, management, membership, and service as well as
financial coverage – in distinctive settings and designed for different population groupsxx. In
theory, there are five characteristics that CHI schemes all sharexxi:
1. Community-based social dynamics and risk pooling, where the schemes are
organized by and for individuals who share common characteristics (geographical,
occupational, ethnic, religious, gender etc.);
2. Solidarity, where risk sharing is as inclusive as possible within a given community
and membership premiums are independent of individual health risks;
3. Participatory decision-making and management;
4. Nonprofit character;
5. Voluntary affiliation.
The nonprofit principle, the premium calculation independent of individual risk and
participatory decision-making clearly distinguish CHI from commercial health insurance,
with which it shares voluntary affiliation. Participatory decision-making, community-based
risk pooling, (usually) flat membership premiums, and voluntary affiliation distinguish CHI
from social health insurance (SHI), with which it shares the nonprofit character. In practice
however, CHI schemes apply the five mentioned principles to a greater or lesser extent.
Schemes set up by healthcare providers, for example might not permit the full development
of participatory decision-making and management. Schemes laid out by government within
a roadmap towards universal coverage might maintain the principle of voluntary affiliation
(in rural Chinaxxii) or make a deliberate choice for mandatory affiliation (in Ghanaxxiii and
Rwandaxxiv).
4
In the anglophone literature, the terms Community Health Insurance and Community-Based
Health Insurance are used most frequently. Less common is the descriptor Mutual Health
Organisation, although its French equivalent Mutuelle de Santé is widely employed in
francophone Africa, thereby emphasising an underlying social dynamic. In West Africa
especially, scheme management relies considerably on community participation. In East
Africa, where provider-driven schemes are encountered more frequently, the financial
dimensions of CHI attract more attentionxxv. This latter approach to CHI is reflected in the
use of the term Health micro-Insurance (HmI). When Dror and Jacquier advanced this concept
back in 1999xxvi, they described HmI as insurance for those excluded from formal social
security (which is often but not always the case in CHI), explained the ‘micro’ in HmI as
lower than national level and admitted that some CHI schemes would fit into the HmI
concept, and claimed that HmI would be less dependent on external subsidies and outside
facilitators than CHI. Especially the latter characteristic is disputable, given the proactive
involvement of international actors and donors over the last decade.
The International Labour Organization endorsed HmI from 2000 on in its programme
Strategies and Tools against social Exclusion and Poverty (ILO-STEP)xxvii, without major
modifications. An influential joint publication on micro-insurance of ILO and the Munich Re
Foundationxxviii replaced “those excluded from formal social security” by “low-income
people”, and explicitly stated two aims of micro-insurance: extending social protection to the
poor and the creation of a new market for commercial insurers, including in the field of HmI.
In the same publication, Radermacher and Dror propose a four-model typology of HmI: a
charitable insurance model (somewhat at odds with the independence claimed seven years
earlier), a community-based model, a provider-driven model, and a partner-agent model
5
(wherein the ‘partner’ can be a commercial insurance company)4. While the first three
models encompass most CHI arrangements, the latter positions HmI much closer to
commercial health insurance than to CHI. Gradual and overt commercialisation of HmI –
and the fact that none of the HmI definitions over time refers to the independence of
premium calculation from individual health risks – refrain us from using HmI as a substitute
for CHI in the framework of universal coverage. Aware that many schemes are classified as
CHI by some and as HmI by others – and that both CHI and HmI are umbrella terms with
many offspring – we use the denominator CHI whenever possible and thereby deliberately
exclude those schemes that do not plainly subscribe to the nonprofit principle.
4 Radermacher R and Dror I (2006) Institutional options for delivering health microinsurance. In: Churchill S [Editor] Protecting the poor: a microinsurance compendium. Geneva: International Labour Office28; pp 401-423.
6
Community Health Insurance in Africa
During the 1970s, the African public health systems deteriorated in parallel with the
deepening economic crisisxxix. From the 1980s on, the introduction of user fees further
impeded access to care and aggravated inequityxxx,xxxi. Community Health Insurance in
Africa must be seen in the context of large majorities within the population trapped in
poverty and excluded from formal social security systemsxxxii. The African CHI movement
was started out of a concern to either improve access to healthcare for a greater proportion of
the population, or to ensure a stable source of income for healthcare provision, or both.
The first initiatives were developed under the direction of expatriate development aid
workers who were most familiar with the history and operation of Europe’s social health
insurance systems. A well-known example is the provider-driven Bwamanda district
hospital scheme in the Democratic Republic of Congo that commenced in 1986 with Belgian
supportxxxiii. Also in 1986, the first community-based schemes emerged with the
inauguration of the Mutuelle Pharmaceutique de la Sainte Famille Tounouma in Burkina Faso.
Over time, different models and blends developed, first in West and Central Africa, followed
later by East Africaxxi.
From the early 1990s on the African CHI movement enjoyed increasing external support,
often from organizations that had a strong attachment to the European SHI model. These
organizations, like for example the International Department of the Belgian Christian
Mutualities, organized training sessions for scheme managers, designed technical manuals
and helped creating and developing local support organisations. Gradually, governments
and donors became interested in the potential of Community Health Insurance to increase
access to healthcare in adverse conditions.
7
It should be noticed however that CHI in Africa also generated critiques, reaching a climax
in the 2008 publication of an Oxfam-led joint NGO briefing paper “Health Insurance in low-
income countries: where is the evidence that it works?”xxxiv.
8
Community Health Insurance in West Africa
In 1998 several African countries, international partners and local actors met in Abidjan to
create the network La Concertation entre les acteurs du développement des mutuelles de santé en
Afrique, known and referred to as La Concertation. This network supports and monitors the
development of CHI, in francophone African countriesxxxv. Its 2004 inventory documented a
nearly five-fold increase in functional Mutuelles de santé between 1997 and 2003 (from 76 to
366 schemes, of which 348 were CHI schemes) in West Africa alonexxxvi. In 2007, La
Concertation changed the methodology of its inventory, which does not allow comparison
between pre-2004 and post-2007 figures. Independent researchers estimated the number of
functional CHI schemes to have grown to 626 by 2006, an eight-fold increase since 1997xxxvii.
This boost in number of schemes in French-speaking West Africa should not detract from the
fact that the bulk of the schemes have less than 1,000 members each. Moreover, most of them
remain firmly linked to a single social setting, such as a village, a neighbourhood or a
professional body. These features lead to a high transaction cost and limited risk pooling
with insufficient an unsustainable coverage of expensive risks, such as surgical interventions
or treatment of chronic diseases.
On a more positive note, West African CHI schemes increasingly aim at improving their
organisational practice by engaging in networks, federations, and unionsxxxviii. And from an
institutional point of view, the West African Economic and Monetary Union’s (UEMOA,
Union Economique et Monétaire Ouest Africaine) call for a legislative framework in every West
African country is promisingxxxix,xl.
9
A closer look at country level gives a more in-depth picture, disclosing different speeds of
implementation, variations in the mode of implementation, and heterogeneous
achievements.
10
Senegal
Based on the number of schemes, Senegal is where the West African CHI movement gained
and maintained its strength. Senegal was home to one out of four West African CHI schemes
in 1997, and to one out of five in 2003 and 2006xxxvii. While most of the earlier schemes started
in rural environments, CHI today is widespread in rural and urban settings. Innovative
initiatives are emerging, such as complementary arrangements to poorly performing
mandatory schemes for formal employees and – since 2007 – a scheme for school children
(AMEL, Assurance Maladie des Elèves)xli, reaching 20,000 enrollees to date.
In terms of financial and population coverage, gains at country level are sensible, and
surpass those of most African countries: the share of prepayment plans (including CHI, not
including mandatory health insurance for formal employees) in private health expenditure
in Senegal rose from 7.1 to 17.9% between 2000 and 2007, while out-of-pocket expenditure as
a proportion of private health expenditure dropped from 91.7 to 78.5% over the same
periodxlii. Still, less than 20% of all Senegalese had any form of contributory health insurance
in 2007, with 4% or less of the population being beneficiary of one of an estimated 130 CHI
schemesxxxvii, ,xliii 5. In the Thiès region, where CHI got first and firmly established, household
survey data from the year 2000 indicated that CHI membership both increased access to
hospital care and decreased out-of-pocket spendingxliv. A 2004 household survey in the same
region confirmed the protective effect of CHI membership on out-of-pocket spending in
inpatient care, but found no significant effect on out-of-pocket spending in outpatient carexlv.
5 Both 4% and 130 are estimates, not counts. The estimated 4% national CHI coverage in 2007 is mentioned in a 2008 BTC/CTB report43, which refers to approximately 420,000 beneficiaries but does not provide traceable substantiation. A 2008 article49 mentions 2.4% national coverage, without stating in which year or substantiating this claim. The 2004 Strategic Plan46 estimates CHI beneficiaries at 421,670 (possibly a source for the estimate from BTC/CTB four years later), but also specifies that only 122,970 (1.2% of the population) of them belong to population groups excluded from mandatory insurance. The estimation of 130 schemes comes from a 2007 overview article37. If the average number of beneficiaries per scheme did not increase (as stated in the 2008 report43), 130 schemes would have matched 3% national CHI coverage.
11
In terms of organisational practice and institutional design, Senegal’s CHI progress has
been slow. Despite the relatively small size of the country, CHI networks exist at regional but
hardly at national level6. Support activities from multiple bilateral agencies follow a similar
regional patternxlvi,xlvii. The constitution in 1998 of a support cell at ministerial level boosted
the creation of new schemesxlviii but had few consequences for design and functioning of the
schemes. Such is also the case with a support cell created within the Ministry of Health and
Prevention for the implementation of the 2004 strategic plan (Plan stratégique de développement
des mutuelles de santé). A legal framework for CHI was established through the 2003 Loi
relative aux mutuelles de santé, but this law is under revision and still lacks an implementation
act. On a more positive note, CHI is increasingly seen by Senegalese policy-makers as one
among a range of measures for social protection and healthcare coveragexlix, such as removal
of user fees for particular services (childbirth), treatments (antiretroviral and tuberculostatic
drugs) and population groups (the elderly, through Plan Sésame)xliii. It should be noticed
however that all these initiatives are experiencing organisational difficultiesxlix, , ,l li lii.
In terms of empowerment, evidence from Senegal is fragmentary and goes in opposite
directions for different vulnerable groups. Data from the already mentioned 2004 household
survey in the Thiès region show a positive association between CHI enrolment and female-
headed households, and between CHI enrolment and institutional deliveries (which led the
researchers to see a role for CHI in the empowerment of women), but a negative association
between CHI enrolment and poverty (also within female-headed households)xlv,liii.
Interesting – definitely when framing empowerment as a step towards a compact of
solidarity – are the findings of a 2002 study: Senegalese CHI promoters valued financial
6 At national level, a forum of exchange exists (CNC, Cadre National de Concertation des acteurs du développement des mutuelles de santé) for donor, CHI and state representatives. Up to today – limiting itself to exchange of information – the CNC cannot be considered a CHI umbrella organisation19.
12
sustainability of their schemes over solidarity; members were split between financial
sustainability and solidarity. Among the members, men and respondents from families with
greater health needs inclined towards solidarity; women towards financial sustainability of
the schemeliv.
Mali
In Mali – with an estimated 102 schemes in 2006xxxvii – the CHI movement benefited from a
permissive legal framework since 1996 and from the existence of a national umbrella
organisation (UTM, Union Technique de la Mutualité Malienne) since 1998.
Gains in terms of financial and population coverage are of little account at country level:
the share of prepayment plans (including CHI, not including mandatory health insurance) in
private health expenditure rose from 0.1 to 0.5% between 2000 and 2007. Yet out-of-pocket
expenditure as a proportion of private health expenditure rose from 99.1 to 99.5% over the
same periodxlii. According to a 2004 report, less than 10% of all Malians had any form of
contributory health insurance, with 0.31% of the population being beneficiary of one of an
estimated 51 CHI schemes in 2003xxxvii,lv.
Persistent low national coverage of social health protection led the Malian government to
elaborate a 2005-2009 action plan7. Among the targets were the introduction of mandatory
health insurance for government and formal sector employees (AMO, Assurance Maladie
Obligatoire), of social assistance for the extreme poor (FAM, Fonds d’Assistance Médicale), and
achievement of 3% CHI coverage at country levellv,lvi. In 2006, national CHI coverage was
7 Announced in 2004 and initially called 2004-2008 action plan for extension of social health protection55; partially operational since 200552, and since referred to as 2005-2009 action plan for extension of social health protection58.
13
estimated either at 0.29%lvi or at 0.33%lvii,8. In 2009 – with AMO and FAM still
nonoperational and national CHI coverage still low – the Malian government launched a
2010-2014 action plan for extension of social health protection, with the same targets as the
former one: introduction of AMO and FAM (now called RAMED, Régime d’Assistance
Médicale) and achievement of 3% CHI coverage at national levellviii.
Coverage by Community Health Insurance at the level of schemes and specific social settings
provides a different story, with divergent characteristics and outcomes. One article based on
a 2004 survey in two rural and two urban Malian CHI settings found evidence for overall
gains in financial coverage – CHI members spending less of their income on health than non-
members – without quantifying these gains nor differentiating them per contextual
settinglix,lx. Another article based on the same survey quantified the gains but found no
significant difference, in outpatient carexlv. Yet another article based on the same survey
observed CHI members spending slightly more for childbirth care than non-membersliii. A
2005 independent evaluation of another CHI scheme found CHI members having a total
direct cost of outpatient care twice as high as non-memberslxi. The variability in these
findings confirms the key role played by contextual dimensions in determining outcomes of
CHI, as highlighted before by Criel and colleagueslxii. In the aforementioned five cases with
five different contexts, the schemes differed not only in financial coverage but also in
population coverage (ranging from 3.1 to 11.4% in the schemes included in the 2004 surveylix;
11.7% in the scheme evaluated in 2005lxi) and in service coverage.
8 All three estimates (0.31% in 2003; 0.29% and 0.33% in 2006) of CHI coverage at country level are either from or stem from Malian authors, who referred to 34,00055, 35,00056 and 40,00057 CHI beneficiaries respectively. External authors have referred to 469,81537 and 499,85636 beneficiaries in 2003, which would correspond to 4.3 and 4.6% national CHI coverage respectively. A 2009 joint Government of Mali/UNICEF publication mentions 100,000 CHI beneficiaries – which would correspond to 0.79% national CHI coverage – but estimates national CHI coverage at 2%58.
14
While these cases seem to indicate that CHI can push coverage in desired and undesired
directions – or just leave it untouched – a sixth Malian case illustrates how a favourable
configuration of intervention characteristics and context features can positively and
sustainably impact all coverage dimensions. The health area of Nongon – comprising
Nongon and neighbouring villages in the southern Sikasso region – is home to Mali’s first
rural CHI scheme, MUTCO (Mutuelle Cotonnière de Nongon). In 1994 the villagers seized the
opportunity created by a recent move towards decentralised governmentlxiii. Organised
through a local unit of the national cotton company9, they constituted a community health
board (ASaCo, Association de Santé Communautairelxiii,lxiv) and built a community health centre
(CSCom, Centre de Santé Communautairelxiii,lxiv), using a 33% levy on cotton saleslxv. From the
start, Nongon’s CSCom has been directed by medical doctors from the Malian rural doctors’
movement (Association des Médecins de Campagnelxvi) with a low turnover rate (1994-2000,
2000-2008, 2009-today), both conditions favouring optimisation of the quality of care. It was
the CSCom’s first director who put forward the creation of a CHI schemelxv, of which he and
his successors have remained active promoters. When MUTCO started in 1998, it was
exceptional in making membership premiums proportional to cotton production, and in
granting the extreme poor free membershiplxvii. Only in 2005 – at the height of the African
cotton crisis – MUTCO shifted to flat-rate membership premiums.
What changes can be observed in population coverage, benefits coverage and financial
coverage? One specific aim of the MUTCO initiators was to increase the utilisation of the
CSCom – 0.35 new cases per person in outpatient care – which they considered inadequatelxv,
even if higher than regional average. Ten years later, outpatient-care utilisation had
9 Respectively ZAER (Zone d’Animation et d’Expression Rurale) and CMDT (Compagnie Malienne pour le Développement des Textiles). Nongon’s ZAER formally established the ASaCo; its boundaries became those of the CSCom’s area of responsibility65.
15
increased to 0.810 – an exceptional achievement in rural Mali, and sub-Saharan Africa.
Underpinning this gain in access was a steady growth in CHI coverage reaching a
penetration rate of 63% in 2004lxvii, and maintaining an enviable 47% in conditions of
widespread poverty10.
One would not expect equal gains in service coverage: the Nongon area has no higher-level
service structure than a CSCom, and the responsibility of a CSCom is to deliver a nationally
circumscribed minimal package of activitieslxiv. Yet collective action of Nongon’s ASaCo,
MUTCO and CSCom director enabled a functioning ambulance service for those in need of
referral care, and decentralisation of services needed but not included in the standard
package – most noteworthy screening and therapy for tuberculosis.
As far as financial coverage is concerned, out-of-pocket spending in outpatient care is
significantly lower for CHI beneficiaries than for non-beneficiaries: four times lower for
pregnant women and children up to age seven, two times lower for other users10. Within the
particular setting of Nongon, all three dimension of coverage increased – including a breadth
of coverage unparalleled in West Africa – and remained relatively high despite economic
hardship.
In terms of organisational practice and institutional design, it is good to remember that
Mali’s 1996 mutual association law (Régissant la mutualité en République du Mali) was the first
of its kind in francophone Africa. Contrary to what happened later in Senegal, the law –
followed by two implementation acts in 1996 and one decree in 1997 – became operational
without delaylvi,lxviii.
10 Own calculations based on primary data from 2006, collected during a 2007 ITM/UTM study on CHI and quality of care in Mali.
16
From 1998 on, the Union Technique de la Mutualité Malienne (UTM) has been the main player
within this legal framework. As an organisation it is one of a kind, exhibiting an amalgam of
characteristics and functions. Besides being created in response to a government call to
French development actors for promotion of CHI, it is not a government agency: it belongs
to the CHI schemes who constitute it, yet is heavily dependent on external financing. Besides
representing the bulk of the Malian CHI schemes in their relation with the government and
the outside world, it gives the schemes technical support for tailor-made CHI suited to local
needs, but also markets its own standard insurance package AMV (Assurance Maladie
Volontaire). And to some degree it fills in for the government’s deficient Directorate of Social
Protection (DPS, Direction de la protection sociale et de l’économie solidaire) in supervising its
own memberslvii,lviii,lxvii,lxviii. In fact, UTM’s multifaceted nature and sizable external financing
have been major forces and have given a voice to the CHI movement, greater than its
numerical strength. By the same token, UTM’s AMV – designed in close cooperation with
the Mutualité Française, looked upon with criticism by other West African CHI promoters, yet
accounting for 60% of all Malian CHI beneficiaries in 2005lvii – more than only broadening
the base for cross-subsidies, brought aboard vocal actors in modern Malian society, such as
independent professionals and government employees.
Not surprisingly, when the Malian government elaborated its 2005-2009 action plan for
extension in social health protection, UTM was the government’s privileged partner. Over
the next years UTM profiled itself as a serious candidate for implementing the mandatory-
insurance component (AMO) of the action plan, for both technical and strategic
reasonsxlviii,lvi,lxvii,11. In the same few years, UTM’s dependence on external financing – the
latter contributing to its strength for almost a decade – became a weakness when one of its
11 UTM repeatedly argued to be the only Malian actor with the sufficient professional capacity to handle the AMO. At the same time the AMO would reduce the demand for the AMV, UTM’s main success story. So UTM had little other option than trying to become a central part of the future AMO.
17
main French donors withdrew. In the subsequent 2010-2014 action plan, the government
restricted UTM’s role to that of increasing CHI coverage to 3% in the informal sector for
those who can pay for it. Supported by the Mutualité Française, UTM reacted by inviting the
minister of Social Development to a seminar on the role of CHI in the extension of social
health protection. Times had changed: after listening to UTM’s renewed appeal for a more
substantial involvement of CHI (and thus UTM) in the new action planlxix, the minister drily
replied by inviting UTM to cover not only the planned 3% but also the remaining 80% of the
population with CHIlxx.
Letourmyxlviii – building on a conceptual distinction made by McIntyre and colleagueslxxi –
describes the Malian government’s approach to achieve universal coverage through health
insurance as an example of a fragmented strategy, as opposed to a more comprehensive
strategy relying on mandatory insurance for the whole population12. With Mali defining its
roadmap based on a separation of the population in three parts – aiming at mandatory
insurance for the formal sector, voluntary insurance for groups able to pay for it in the
informal sector, and social assistance for the extreme poorxlviii,lviii – UTM today has the
difficult task to reinvent its organisational practice within an institutional design much less
favourable for CHI than it was a decade ago.
In terms of empowerment and in addition to the changing picture at national level,
promising evidence at household and scheme level can be found in the case of Nongon
presented above. In Nongon, a majority of women indicated that being a beneficiary of the
CHI scheme had made irrelevant the authorisation of their husbands to seek consultation.
According to the Nongon women, they now had autonomy of decision for which no more
12 Letourmy (2010) gives the Ghanaian and Rwandan health insurance reforms in process as examples of a comprehensive strategy48. See further for a more in-depth discussion of the Ghanaian and Rwandan reforms.
18
than pocket money was neededlxxii. The importance of such impact cannot be
underestimated in a country were only 18% of women in need of care receive their
husband’s authorisation to go for a consultation, compared to 53% who have the money to
do solxxiii.
Guinea
Guinea – with an estimated 90 schemes in 2006 – comes close to Mali in number of schemes,
but with significantly smaller schemesxxxvii. The Guinean CHI movement presents two
particularities: a research project that set out the stakeslxxiv, and a series of schemes targeting
pregnant womenlxxv.
Financial and population coverage at country level leaves much to be desired. No gains are
apparent: the share of prepayment plans (including CHI, not including mandatory health
insurance) in private health expenditure was reported 0% in 2000 and 2007; out-of-pocket
expenditure as a proportion of private health expenditure remained at 99.5%xlii. Less than
1.5% of Guineans were covered by CHI in 2006xxxvii.
Organisational practice and institutional design present likewise weaknesses. Guinea has
no legal framework for CHI; no national umbrella organisation existslxxvi. In one of Guinea’s
four regions a CHI coordination structure is in place: the Union des mutuelles de santé de la
Guinée forestière, which brings together 25 schemes with a total of about 14,000
beneficiarieslxxvii. It was in the same region that the research project PRIMA (Projet de
Recherche sur le Partage du Risque Maladie) developed and tested a model of rural CHI
(MUCAS, Mutuelle Communautaire d’Aire de Santé) between 1996 and 2000lxxiv. From Dabola
district in the neighbouring Haute Guinée, another model emerged: CHI schemes for safe
19
motherhood (MURIGAs, Mutuelles pour la prise en charge des Risques liés à la Grosesse et à
l’Accouchement). Rolled out by the Guinean government with support of UNICEF, MURIGAs
in 2006 covered 10% (about 16,000 women) of their target population in 17 out of 33 health
districts. In a handful of districts children were added to the MURIGA’s target group, with
little effect so far. Management of the schemes is substandard and community involvement
is minimal, the latter despite the fact that MURIGA’s were intended to be a community
participation mechanismlxxv.
In terms of empowerment, both promising and discouraging evidence can be found at the
level of the healthcare users/providers interface in the case of the Maliando scheme, in the
area of Yendé. The Maliando scheme was the first realisation of the MUCAS model, which
among other objectives aimed at improved service quality through a CHI-induced
partnership between the community and the providerslxxiv. Patients arrived at a good
understanding of CHI as an insurance mechanism. They also gained voice through CHI and
started claiming their right to good quality carexviii,lxxviii. Yet, contrary to what happened in
the Nongon case (Mali, see above) where the doctor of the health centre became the CHI
scheme’s main activist, several local providers in Yendé did not internalise the partnership
concept and appealed for a cap on CHI membershiplxxix, ,lxxx lxxxi. Suggested explanations for
the providers’ reservation are a conflict between dominant medical culture and the need for
transparency and accountability induced by CHI, and – in this particular case – the fact that
up to 60% of the health centre’s profits were based on informal paymentslxxx,lxxxii.
20
Burkina Faso
Burkina Faso had an estimated 60 schemes in 2006, most of minor sizexxxvii. One of them is
the CHI scheme/research project of Nouna district – set up in 2004 – where determinants of
membership and effect on healthcare utilisation have been extensively studied.
Gains in terms of financial, population and service coverage are marginal at country level:
the share of prepayment plans (including CHI, not mandatory health insurance) in private
health expenditure rose from 1.0 to 2.0% between 2000 and 2007, while out-of-pocket
expenditure as a proportion of private health expenditure dropped from 94.4 to 91.3% over
the same periodxlii. Less than 0.2% of Burkinans were covered by CHI in 2006xxxvii. Coverage
of the target population in the Nouna scheme reached 6%, but was plagued by dropout rates
as high as 45%, mainly associated with affordability issues and perceived low quality of
carelxxxiii. When quality of care was perceived good, financial barriers might still have
prevented people to enrol. In a context of manifest underutilisation, CHI membership
significantly increased outpatient – but not inpatient – healthcare utilisationlxxxiv.
In terms of organisational practice and institutional design, state and civil society actors
move at different speeds: Burkina Faso has still no legal framework for CHI; yet a third-party
CHI support organisation exists since 1999 (RAMS, Réseau d’Appui aux Mutuelles de Santé du
Burkina Faso). Together with a national labour union and microfinance cooperative, plus a
regional development organisation, RAMS constitutes a platform for the Burkinan social
economy (PARESOC Burkina Faso, Programme d’action régionale pour l’économie sociale)lxxxv,13.
13 PARESOC platforms – supported by the Belgian Survival Fund (BOF, Belgisch Overlevingsfonds; FBS, le Fonds Belge de Survie) – exist in Benin, Burkina Faso and Mali.
21
In terms of empowerment, evidence from the Nouna project highlights inequity in both CHI
enrolment and healthcare utilisation: the very poor were less likely to enrol; once enrolled,
they were less likely to utilise health services compared to their wealthier counterparts14.
This led the researchers to the conclusions that CHI enrolment of the very poor needs to be
subsidised, and that complementary measures are needed to enhance the capacity of the
deprived to make use of the serviceslxxxiv. From the Burkinan PARESOC initiative – which
included from the start complementary measures within a broader development perspective
– no evidence on empowerment is yet documented.
Benin
Benin had an estimated 120 schemes in 2006xxxvii. Benin’s CHI development shares several
characteristics with that of Senegal.
Financial and population coverage are modest at country level: the share of prepayment
plans (including CHI, not mandatory health insurance) in private health expenditure rose
from 0.1 to 5.0% between 2000 and 2007, while out-of-pocket expenditure as a proportion of
private health expenditure dropped from 99.9 to 94.9% over the same periodxlii. Less than
1.4% of Beninese were covered by CHI in 2006. In terms of numbers of schemes, Benin is
West Africa’s fastest grower after Senegalxxxvii.
Organisational practice and institutional design hardly keep pace with the proliferation of
schemes. Benin has no specific legal framework for CHI, although a CHI code is in the
14 Such finding is of course not new, but consistent with a course of action that Julian Tudor Hart had already identified back in 1971 in the British National Health System and led him to formulate his inverse care law: “The availability of good medical care tends to vary inversely with the need for it in the population served”. See: The Lancet 297(7696), 405-412.
22
making since 2005xix. As in Senegal, a handful of regional networks exist – each supported by
a different donor – but no national umbrella organisationxix,15.
In terms of empowerment, evidence is mixed. Research on four CHI schemes within the
Réseau Alliance Santé (RAS)16 reported that members had become aware of their right to
complain as healthcare users – a privilege once enjoyed by the wealthy only – and
highlighted the role of CHI to negotiate good quality of carelxxxvi. A visit to the five schemes
of the Union Communale des Mutuelles de Santé de Bembèrèkè17 returned with the message that
networking had enabled dialogue with the healthcare providers, which among other things
had reduced cost of carelxxxvii. Yet both sources also mentioned non-inclusion of the extreme
poor as an equity issue. For the managers of the RAS schemes, this was not a matter of
concern: they were mainly interested in the financial viability of the schemeslxxxvi. For the
managers of the Bembèrèkè scheme, it was an unresolved problem: they had not been
successful in claiming subsidies from national social assistance funds, and were financially
unable to constitute a social assistance fund by themselveslxxxvii.
15 At national level, a forum of exchange between the support structures is formally in place (CONSAMUS, Concertation Nationale des Structures d’Appui aux Mutuelles de Santé). 16 The Réseau Alliance Santé (RAS) – assisted by the French CIDR, Centre International de Développement et de Recherche) – is a regional Beninese CHI network, comprising 29 schemes. 17 The Union Communale des Mutuelles de Santé de Bembèrèké is a communal CHI network, assisted by the Belgian PARESOC and PROMUSAF initiatives, the latter giving support to a total of 20 schemes.
23
Togo
Togo is a late developer in CHI, its schemes appearing from 2000 on. In 2006, Togo had an
estimated 12 schemesxxxvii.
Gains in terms of financial and population coverage are marginal at country level: out-of-
pocket expenditure as a proportion of private health expenditure dropped from 86.6 to 84.2%
between 2000 and 2007, while the share of prepayment plans (including CHI, not including
mandatory health insurance) in private health expenditure actually dropped from 5.4 to 4.3%
over the same periodxlii, ,lxxxviii 18. Less than 0.5% of Togolese were covered by CHI in 2006xxxvii.
At scheme level, progress in depth and height of coverage are noted: understanding and
applying the principles of risk spreading, most schemes now include C-sections in their
benefit package. Younger schemes do this without imposing a ceiling; one scheme is
including emergency care for children with malaria as a second felt needlxxxix.
Organisational practice and institutional design shows little development at national level.
Togo has no legal framework for CHI; no national umbrella organisation exists. So far, no
evidence of empowerment is at hand.
18 It should be remembered that Togo (together with Gabon) is one of the few West African countries were mandatory health insurance has not considerably eroded post-independence, maintaining 23% population coverage (29% in Gabon)84.
24
Cameroon
Cameroon was home to a slow but steady CHI development since the 1990s and reached an
estimated 30 schemes in 2006xxxvii. Much has happened since. Intense promotion of CHI by
both government and development partners led to an incremental growth in number of
schemes, reaching 107 schemes in 2008xix,xc. Enrolment rates are much less impressive.
Gains in financial and population coverage at country level are not apparent, at least not
before 2008: the share of prepayment plans (including CHI, not including mandatory health
insurance) in private health expenditure was reported 0% in 2000 an 2007; out-of-pocket
expenditure as a proportion of private health expenditure remained at 94.5%xlii. Less than
0.2% of Cameroonians were covered by CHI in 2006xxxvii,19. In 2007, five schemes piloted an
effort to extend benefits coverage with HIV/AIDS care, with support of the German
Technical Cooperation (GTZ)xci. The Cameroonian government has declared covering
expensive risks a priority and wants to achieve this by establishing reinsurance funds, but
financing is still lackingxix.
Organisational practice and institutional design are scaling up relatively fast in Cameroon.
Despite not having a specific legal framework for CHI, Cameroon has included CHI
development explicitly in its 2001 and 2006 strategic plans. The 2001 strategic plan aimed at
one CHI scheme per health district in 2010, and 40% population coverage by the same. These
goals were not attained. The redressed 2006 strategic plan still aims at one CHI scheme per
health district – now to be achieved by 2015 – and either 40% or a possibly more realistic 20%
19 It would obviously be interesting to have data on CHI population coverage following the boom in schemes after 2006. To our knowledge, such data are not yet available.
25
population coverage by the same year20. Six donor-assisted and/or regional CHI networks
are operating, but no national umbrella organisation is in place. Yet, in 2006 – in addition to
the national forum of exchange (CNC) existing since a year earlier – a platform21 of 39 local,
regional and national CHI promoters was established, in partnership with the Ministry of
Public Health, the Ministry of Employment and Social Protection, and a range of bilateral
and multilateral cooperation agenciesxix,xc,xci. As of today, the accent is on the development of
schemes, in number and coverage. Problems in quality at the supply side – a hindrance itself
for CHI enrolment as frequently mentioned by local actors – are much less addressed.
So far, no evidence of empowerment is at hand.
Niger
Niger is a late and slow developer in CHI, its schemes appearing from 2000 on. In 2006,
Niger had an estimated 18 schemesxxxvii; in 2008, 17 schemes were countedxix. Extreme
poverty and low government capacity encumber any development, including that of CHI.
Financial and population coverage at country level leave much to be desired. Niger in fact is
loosing ground: the share of prepayment plans (including CHI, not including mandatory
health insurance) in private health expenditure dropped from 11.3 to 3.2% between 2000 and
2007; out-of-pocket expenditure as a proportion of private health expenditure increased from
87.6 to 96.4% over the same periodxlii. Less than 0.7% of Nigeriens were covered by CHI in
2006xxxvii; depth of coverage is minimalxix.
20 Unclear if the actual aim was 20 or 40%: 20% according to the francophone communication, 40% according to the anglophone communication. Compare http://www.plateformecm.org/plateforme/index2.php?cat=planstrategique and http://www.plateformecm.org/plateforme/en/index2.php?cat=strategicplan21 Promuscam (Plate-forme des Promoteurs de Mutuelles de Santé au Cameroun).
26
Similarly, organisational practice and institutional design show little substantial
development. No national umbrella organisation exists; one NGO-based network operates
since 2002. A national forum of exchange (CNC) exists but shows little activity. A 2008 law
advocates CHI development and expansion; the Ministry of Civil Services and Labour would
be responsible for supervision and regulation, a support unit within the Ministry of Public
Health for promotion of CHI. The law has no implementation act; the institutions lack
financing and capacityxix.
No evidence on empowerment is yet documented.
Mauritania
Financial and population coverage at country level leave much to be desired. No gains are
apparent: the share of prepayment plans (including CHI, not including mandatory health
insurance) in private health expenditure was reported 0% in 2000 an 2007; out-of-pocket
expenditure as a proportion of private health expenditure remained at 100%xlii.
These worrisome data at country level are reflected at scheme level. Of the five schemes
active in 2006xxxvii, and three in 2003xxi,xxxvi, only one still exists: the Mutuelle Communautaire de
Santé de Dar-Naïm (MCSDN). The scheme was conceived in 2002 and started in 2003 in a
poor neighbourhood of the capital (Dar-Naïm, Nouakchott) within an existing primary
health project (PSDN, Projet Santé de Dar-Naïm) of Caritas Mauritaniaxcii. The uniqueness of
MCSDN is not only on account of being the only one left; it is also merited by the scheme’s
27
novel and successful combination with a health equity fund (HEF22, Fonds d’indigence de Dar-
Naïm), since 2005xciii.
Financial, service and population coverage of the MCSDN is documented and deserves a
closer look. Financial coverage is limited and an intricate set of copayments is in place23.
Service coverage has been gradually expanded since 2003 and is today comprehensive as far
as primary care is concerned, but still limited at referral level. Population coverage never
reached 5% of the target population and converts the Dar-Naïm scheme in a showcase for
the limits of CHI design (community-based and with voluntary affiliation) in a specific
context (a society consisting of segregated social groups). Whatever efforts the MCSDN has
made to grow, there seems to be no way to overcome the existing geographical and social
boundaries. As a result, risk pooling remains equally limited, and further expansion of
financial and service coverage is extremely difficultxciii,24.
Institutional design deserves few comments: Mauritania has no consistent CHI framework
in place. The government operates a Fonds d’indigence at national level, but leakage is the
rule: beneficiaries are often privileged, not destitute, members of society.
Operational practice becomes interesting at Dar Naïm level, and especially so in the setup
and implementation of MCSDN’s health equity fund. At onset, identification of beneficiaries
of the HEF was based on a list of criteria and resulted in the selection of a narrow group of
22 Health equity funds (HEF) – social assistance mechanisms based on third-party financing to guarantee the extreme poor access to healthcare – originated in Cambodia in 2000. For a discussion of HEF in Cambodia, see the section on Cambodia in this paper and the accompanying references: Hardeman and colleagues (2004)222, Jacobs and Price (2006)223, Noirhomme and colleagues (2007)224, Annear and colleagues (2008)232, Meessen and colleagues (2008)225, and Bigdeli and Annear (2009)226. The Fonds d’indigence de Dar-Naïm was not the first HEF in Mauritania or Africa. In Mauritania, the HEF of Hodh el Gharbi and Hodh el Chargui started in 200393. But is has been the most successful, displays innovative design and practice, and subsists to date. 23 Comprising a flat copayment for outpatient care, and proportional copayments ranging from 25% for delivery services to 70% for drugs for chronic diseases93. 24 External evaluators of the MCSDN have suggested the creation of new schemes as an alternative for scheme expansion. It can be noted that CHI in the Thiès region (Senegal) progressed along similar lines. From a pragmatic point of view, this can be considered a plausible way forward (though no guarantee for improved risk pooling, for which then networking could be a solution). From a conceptual point of view, this raises a question mark over the feasibility of voluntary solidarity.
28
permanently destitute. Over time – and in joint agreement between scheme management and
the community – the programme has been extended to include the temporary poor, to
prevent them from falling into destitution25. Concurrently, the service coverage for the HEF
beneficiaries has extended in two ways to respond to their specific needs: it covers a more
comprehensive package of medical care than that proposed to CHI contributors, and it
includes social services beyond the medical sphere. One more novelty is worth noticing: a
HEF staff member is responsible for house visits and assessment of vulnerability of potential
and actual beneficiaries. Such approach opens room for better understanding of needs and
appropriate action26,xciv.
Evidence on empowerment is limited to the boundaries of Dar Naïm and its health equity
fund, where the assistance provided indeed seems to engender inclusion and autonomy
among the most vulnerablexciii.
Ghana
In anglophone West Africa, the case of Ghana is of particular interest. Since the introduction
of user fees – called ‘cash and carry’ in Ghana – in the mid 1980s, both government and
private actors have been exploring the feasibility of health insurance.
After lengthy consultations, the Ghanaian government piloted (and immediately
abandoned) a National Health Insurance Scheme in 1997 and a Ghana Health Care Company
in 1999. Already in 1989, a faith-based nonprofit service provider – inspired by the
25 Coverage by the MCSDN’s HEF shows the following evolution over time: 51 persons in 2005; 200 in 2006; 293 in 2007 and 401 in 2008 – corresponding to 0,1 (2005); 0.3 (2006); 0.5 (2007) and 0.7% of Dar Naïm’s population93. 26 Both inclusion of non-medical services and the engagement of a skilled social worker are novelties in HEF practice. They bear similarities with effective modern social assistance practice in Western Europe94.
29
Congolese Bwamanda scheme27 – had designed the Nkoranza scheme, which became
operative in 1992. Other schemes followed, either community-initiated or provider-initiated
as Nkoranza, most them arising after a CHI workshop organised by PHR28 in 1999. By 2001,
14 CHI schemes were fully functional. Yet most of them were small29 and together they
covered only a fraction of the population.
Also in 2001, a new government reaffirmed the aim to replace cash and carry by a National
Health Insurance Scheme (NHIS)30. This led to the 2003 National Health Insurance Act
(NHIA), which significantly changed health financing across the country. By prescribing
every district to set up a CHI scheme31, the NHIA paved the way for scaling up community
health insurance within the NHIS, a social health insurance construct aiming at universal
coverage in the long run. The NHIS became operational from 2005 on. By the end of 2005, 83
out of 138 districts boasted a CHI scheme, covering between 20 and 40% of their
populationxxiii, , , , ,xcv xcvi xcvii xcviii 32.
Gains in terms of financial coverage at country level were not apparent by 2007, in
contradiction with the gains in population coverage of (in principle) mandatory health
insurance: out-of-pocket expenditure as a proportion of private health expenditure hardly
27 In the DR Congo – or Zaire between 1971 and 1997 – the Bwamanda hospital-based scheme is active since 1986. See pp 57-58. 28 Partners for Health Reform, USAID project for health policy and health systems strengthening, later PHRPlus (Partners for Health Reformplus). See http://www.healthsystems2020.org/content/resource/detail/670/29 The big exception being the Ghanaian CHI pioneer Nkoranza, which covered 48,000 people (30% of its target population) in 200195. 30 Several authors explain the political context for the reappearance of the NHIS concept in 2001. Ghana has elections every four years since 1992; the last three times being in 2000, 2004 and 2008 – each time with health insurance prominently on the agenda. In 2000, the then oppositional New Patriotic Party (NPP) centred its electoral campaign on the promise of replacing cash and carry by NHIS. The NPP won the 2000 elections and remained in power through the 2004 elections23,98,102,103. In 2008, the National Democratic Congress (NDC) made the introduction of a ‘one-time premium’ for the informal sector’s NHIS contribution a key element of its campaign, regained power and constituted the 2009-2012 government103 (see further in the section on institutional design and organizational practice). 31 In Ghana since called District Wide Mutual Health Insurance schemes (DWMHI). 32 Ghana’s post-2003 insurance reform is the foremost example in Africa (the second one being Rwanda) of what Letourmy calls a comprehensive strategy48.
30
dropped between 2000 and 2007 (from 79.6 to 79.3%)xlii,33, whereas CHI population coverage
as expressed by proportion of cardholders was reported 44% in 2007lxxi, coming from 6.6% in
2005 and reaching one of two Ghanaians in 2008, more than 11 million peoplexxiii,lxxi, ,xcix 34.
Witter and Garshong have further scrutinised available primary and secondary data on
affiliation and utilisation, service coverage, and financial protection within the NHIS’s
district schemes (DWMHIs, District Wide Mutual Health Insurance schemes). Regarding
affiliation, they noticed a pro-rich bias in enrolment and a pro-urban bias in renewal of
membership. They reported utilisation rates of outpatient care in 2006 almost twice as high
for members as for non-members35. They described service coverage as high as 95%36, but
found no conclusive evidence on changes in financial coveragexxiii.
Since the National Insurance Health Act lays down an institutional and financial framework,
institutional design and organisational practice of CHI today are obviously intertwined in
Ghana. The NHIA established the National Health Insurance Council (NHIC, the Council)
and created the National Health Insurance Fund (NHIF, the Fund). The Council accredits
and regulates both district-based CHI and commercial health insurance schemes. It also
manages the Fund, whose main function is to subsidise the district-wide CHI schemes
33 The share of non-mandatory prepayment plans in private health expenditure is no longer a relevant indicator for the impact of CHI in Ghana, since CHI in principle became mandatory in 2003. This share dropped from 6.1 to 5.9% between 2000 and 200742. 34 By December 2007, 55% of the population had registered for and 44% had received a membership card from one of 145 DWHIs (we assume that by then every district had a scheme, 145 exceeding the actual number of 138 districts)71,99. Witter and Garshong report 45% cardholders in 200823. As mentioned before (see page 1) and noted by Mills, inclusion within a financing scheme does not automatically lead to benefits, such as actual access and financial protection7. 35 0.9 consultations/inhabitant/year for members vs. 0.49 for non-members, according to a 2006 ILO study. Utilisation among members further increased to nearly 1.5 consultations/inhabitant/per year in 2009, according to information from the Network of Mutual Health Organizations23. A baseline study at the onset of the NHIS (in Nkoranza and Kwahu South, where CHI schemes were operative since 1992 and 2001 respectively) had also reported utilisation of outpatient care nearly twice as high for scheme members as for non-members96. 36 The service package offered under the Ghanaian NHIS is indeed comprehensive, when compared to service coverage elsewhere and certainly so in West Africa. It includes both primary and referral care, and drugs listed in a National Health Insurance Drug List. Among the main exclusions are cosmetic surgery, transplantations, cancer treatment (other than for cervical and breast cancer), and antiretroviral drugs23. Notwithstanding this considerable depth of coverage foreseen by NHIS, availability of services and manpower disparately restricts access, especially in the poorer northern regions (in 2006 home to 18% of the population; served by 8% of all hospitals and 5% of all doctors)97.
31
(DWHIs)37 that constitute the NHIS. The Fund is financed by a combination of earmarked
levy (NHIL, National Health Insurance Levy, consisting of 2.5% of VAT, value-added tax),
other earmarked government funds, a transfer from the formal sector workers social security
contributions (SSNIT, Social Security and National Insurance Trust, transferring a 2.5%
payroll deduction), and individual premiums for enrolees from the informal
sectorxxiii,xcvii,xcviii,c. Ghana has thus adapted a social health insurance model to facilitate the
inclusion of informal sector workers, by envisaging a network of CHI schemes under a
centralised authority and fundingc.
The NHIS design and its implications have been subject to both political controversy and
scholarly research. In the 2001-2003 pre-legislation phase, formal workers – backed by the
then oppositional National Democratic Congress – strongly opposed cross-subsidising health
insurance for the informal sector with part of their SSNIT contributions. The 2003 NHIA
compensated the formal workers by exempting them from premium payment for the
DWHIsxcviii. Yet a clash of interests between formal and informal workers with regards to the
NHIS is persistentc. The political debate raise again after 2008, when the National Democratic
Congress38 announced the introduction of a ‘one-time premium’ for the informal workers as
a replacement of annual premiums. The debate further complicated when international
actors urged the Ghanaian government to implement the ‘one-time premium’ immediatelyci,
and is still ongoingcii.
37 But not non-district-wide schemes, which could still operate but were classified as private by the NHIA. This was one of the contested issues in the 2001-2003 pre-legislation phase98. 38 The National Democratic Congress (NDC) – which won the 2008 elections – tends to have a stronger following among the poorer northern and urban communities. The New Patriotic Party – which had implemented the NHIS – tends to have a stronger following among the more prosperous southern and urban communities98.
32
Since its implementation in 2005, scholars have voiced equity concerns over NHIS because
the bulk of NHIS’ funding is from VAT39, which is in principle a regressive form of
taxationxxiii. Recent research however shows that VAT is actually progressive in Ghana, as a
range of goods largely consumed by the poor are purposely exempt from this taxcii,40. Equity
in uptake of insurance under NHIS is another issue. Membership is mandatory, but not
enforceable in the informal sectorcii. In a 2008 household survey in one district, Sarpong and
colleagues found 21% of poor households enrolled in NHIS, compared to 60% of those
classified as rich. In addition to greater constraints to pay for the premiums, lack of
consistent information on the NHIS contributed substantially to low coverage among the
poorciii. Confusion over basic details of the NHIS – including the cost of premiums – was
confirmed by Alfers through focus groups with women in the informal sectorc. Jehu-Appiah
and colleagues confirm relatively low uptake of NHIS by the poor, due to implementation
problems and despite the legal provision for premium exemptions for the extreme poorxcix.
According to McIntyre and colleagues, the implementation of these exemptions is likely to
be more inapt in the lowest-income regions, which also have lowest staffing levels and
weakest service deliverylxxi. The Ghanaian policymakers and implementers are aware of
these difficulties. At central level, both the amount and the modality of payment of the
premiums are under revision, as are the conditions for exemption of premium payment41. At
scheme level, efforts have been made for improved identification of the extreme poor.
A number of elements of actual organisational practice in line with the institutional design
merit a short discussion. One already observed strength is the schemes’ comprehensive
39 In 2006, the National Health Insurance Levy amounted to 76% of the National Health Insurance Fund100. 40 See also the 2010 SHIELD information sheet ‘Who pays for health care in Ghana?’ http://web.uct.ac.za/depts/heu//SHIELD/reports/SHIELD_Ghana_WhoPays%20forHealthCare.pdfThis policy brief however warns that VAT in Ghana is verging towards proportionality and could become regressive over time. 41 For example, exemption was extended to all pregnant women (in 2008) and to all children under five (in 2009). The promise was made to replace the informal workers’ annual contributions with a ‘one-time premium’, which led to a wider political debate and is still an unresolved issue – as mentioned before23,102.
33
benefit package. It should be noted however that this attractive feature urgently lacks
improvements at the supply side of care to effectively translate population coverage into
benefits coverage.
From their inception, the DWHIs under NHIS have been operating without copayments at
the time and place of service, congruent with the aim of increasing utilisation. But providers
are still paid on a fee-for-service basis, which is prone to supply-induced demandxcvii. It
should thus come as no surprise that several authors have pointed to the threats of cost
escalation42 and financial instabilityxxiii,xcvi,xcvii,xcix.
Indeed, Ghana – having come a long way – today still faces the difficult balance between the
aim of equitable universalism and the limits of economic efficiencyxcviii,xcix,cii.
Acknowledging Ghana’s explicit aim to make substantive improvements for the poor, the
presence and nature of a specific dynamic of empowerment and transformation would be a
proper subject of monitoring and research, which has barely received attention so far.
In her analysis of the NHIS, Alfers mentions the representation of organised workers by one
member in the National Health Insurance Council, which she considers insufficient given the
fact that one individual has to voice the interest of both formal and informal workers.
Another identified deficiency is representation of the informal workers at scheme levelc.
Witter and Garshong point to a closed corporate culture within the governing bodies of
NHISxxiii.
42 A cost escalation further accentuated by poor gate-keeping in Ghanaian health service delivery (independent from NHIS), according to Witter and Garshong23.
34
Community Health Insurance in East and Central Africa
In East and Central Africa, Community Health Insurance is lately enjoying increased
attention. Both healthcare providers and governments tend to play a prominent role in the
launch and management of CHI schemes. In the United Republic of Tanzania, Kenya and
Uganda most schemes are of recent origin and their numbers remain smallciv. In Rwanda, the
CHI – or Mutuelles – approach became a central government policy in 1999 and achieved
remarkable population coverage in a less than a decade, which we will discuss in detail. In
Burundi and the Democratic Republic of Congo, CHI has a long history (from which still
valid lessons can be learnt), declined over time and is currently re-emerging.
35
The United Republic of Tanzania
In the United Republic of Tanzania, user fees have progressively replaced financing from
general taxation since 1993. From 1996 on, the Tanzanian government introduced
Community Health Fund (CHF) schemes, essentially a district-based CHI arrangement.
Alongside, a dozen provider-driven CHI schemes originated. A 2001 CHF act made the
creation of a Community Health Fund obligatory for every rural district within a two-year
span43 and introduced state subsidies for CHF schemes: member fees are matched by a 100%
government grant.
Gains in terms of financial and population coverage are modest at country level: the share
of prepayment plans (including CHI, not mandatory health insurance) in private health
expenditure rose from 4.5 to 10.4% between 2000 and 2007, while out-of-pocket expenditure
as a proportion of private health expenditure dropped from 83.5 to 75.0% over the same
periodxlii.
Enrolment in CHF schemes remained far below expectations for over a decade. A study
published in 2007 recorded an average enrolment rate of 10%, and identified inability to pay
membership fees, low perceived quality of care and lack of trust in the scheme management
as barriers to enrolmentcv. Where people are enrolled, members utilise services more than
non-members and have better financial protectioncvi,cvii.
Regarding organisational practice, some features of the CHF schemes merit closer attention.
First of all, service coverage in principle includes inpatient and outpatient care at both
43 Five years later, in 2006, 69 out of 92 rural district councils had an operative CHF scheme108,110.
36
dispensaries and first-referral level44. Daily management of a scheme is the responsibility of
the district council, composed largely of government officials. A fraction of the people too
poor to pay a contribution gets free membership. Identification of those too poor to pay is
the responsibility of the respective village councils, whereas the district council is expected to
subsidise the membership fees of the exempted. Health care providers linked to a CHF
scheme receive an advance capitation grant to allow for sufficient equipping of the
facilitycvi,cvii,cviii.
Lately, innovative features were introduced in Tanzania’s Mbeya region: the blending of a
NGO driven community-based scheme with a CHF scheme45, the transfer of CHF
management functions from the district council to a member-based organisation46, and the
use of corporate subsidies to pay for premiums. The latter is an interesting example of the
principle of corporate social responsibility. In the Mbeya’s Rungwe district, the local tea
company subsidises the premium of the tea farmers and workers, which has allowed the
CHF scheme to reach 30,000 beneficiaries (18% of the target population) in two years timecix.
At national level, the Tanzanian Network of Community Health Funds (TNCHF)47 – assisted
by the Tanzanian-German Programme to Support Health of the German development
cooperation (GTZ)48 – provides technical and managerial support since 2003.
Regarding institutional design, the interaction between the National Health Insurance Fund
(NHIF) and CHF is worth noticing. Established in 1999 and operative since 2001, the NHIF is
44 First-referral level being health centres in rural areas of the United Republic of Tanzania. Initially, coverage did not include hospital care. Over time, district councils started including hospital care in their CHF benefit package108. 45 Currently, two community-based scheme are notable in the United Republic of Tanzania: an NGO-driven scheme in Mbeya (about 12,000 beneficiaries) and a provider-driven scheme in Bukoba (about 8,000 beneficiaries). The blend with a CHF scheme allowed the Mbeya scheme to capitalise on the matching grant of the government for every member fee paid. 46 This is a pilot project that aims at improved community participation in local CHF scheme management. 47 See http://www.tnchf.or.tz/48 See http://www.tgpsh.or.tz/
37
a social health insurance scheme for civil servantscx. Currently, the financial management of
the CHF schemes has been centralised within the NHIF – thereby improving the CHF
schemes’ risk pooling function – whereas the daily management of each scheme remains a
district responsibility.
The already mentioned government’s decision of allocating a matching grant equal to
member contributions has been an effective incentive for enrolment in CHF schemes in the
long run. It has been argued however that these matching grants cannot be considered an
equitable allocation mechanism, as relatively poor districts are less able to generate
contributions in the first placelxxi.
Regarding the dynamic of empowerment and transformation, it is worth noticing that the
health financing programme component of GTZ’s TGPSH has empowerment of users – at
individual and community level – as a specific objective and has been piloting several
projects to this end. Results of these efforts still need to be documented.
Kenya
Gains in terms of financial and population coverage are modest at country level: the share
of prepayment plans (including CHI, not mandatory heath insurance) in private health
expenditure rose from 7.1 to 8.8% between 2000 and 2007, while out-of-pocket expenditure
as a proportion of private health expenditure dropped from 80.1 to 77.2% over the same
periodxlii.
38
In Kenya, two features of organisational practice and institutional design are today of
special interest.
On the one hand, the Kenya Community Based Health Financing Association (KCBHFA)49
actively promotes the integration of CHI an its members in the National Hospital Insurance
Fund (NHIF)50 and in its projected successor, the National Social Health Insurance Fund
(NSHIF)51, ,cxi cxii.
On the other hand, the Kenyan microfinance and cooperative movement becomes
increasingly involved in CHI. An example of CHI introduced by a microfinance institution is
that of the Jamii Bora scheme, since 200152. This scheme currently has 91,000 beneficiaries.
An example of CHI grafted on a cooperative movement is that of the Co-operative Insurance
Company (CIC)53, also since 2001, which currently operates 75 schemes with a total of
300,000 beneficiaries.
Uganda
In Uganda, faith-based hospitals started CHI schemes in the late 1990s54. Some community-
based schemes followed55. Many collapsed in 2001 when user fees were abolished in the
public health services. Community health insurance schemes survived where people seek
49 Platform organization for CHI actors in Kenya since 2002. See http://www.kcbhfa.org/50 Social health insurance scheme, since 1966. See http://www.nhif.or.ke/healthinsurance/ 51 Enacted in 2004, but not yet operative. 52 See http://www.jamiibora.org/health.htm53 See http://www.cic.co.ke/Our-Policies/Micro-Insurance54 These provider-driven schemes received support from DFID – the UK Department for International Development – in their start-up phase. DFID also helped establishing a CHI support organisation – UCBHFA, Uganda Community Based Health Financing Association, see http://ucbhfa.org/ – in 1998. UCBHFA largely failed its capacity-building objective, and subsists mainly as a CHI umbrella organisation. 55 These schemes received support from CIDR – Centre International de Développement et de Recherche – in their start-up phase. Their organization design is similar to the community-based schemes commonly found in West Africa.
39
care in private health services, particularly in faith-based hospitals and health centres56. In
2006, 13 schemes were still active in this environmentciv, and their number is slowly growing.
Of the community-based schemes, one is still active today57.
Gains in terms of financial and population coverage are modest at country level: the share
of prepayment plans (including CHI, not mandatory heath insurance) in private health
expenditure rose from 0.1 to 0.2% between 2000 and 2007, while out-of-pocket expenditure
as a proportion of private health expenditure dropped from 56.7 to 51.0% over the same
periodxlii. Population coverage at scheme level is not impressive either. In 2006, the existing
schemes reached on average 8% of their target population. As reasons for not enrolling the
usual suspects were identified: inability to pay the premium, perceived low quality of care,
and lack of trustcxiii,cxiv. Scheme-specific data on service and financial coverage are not
available.
Organisation practice is documented in a 2009 evaluation of 9 CHI schemes in Kabale and
Masaka diocese, southwest Uganda. Following identification of obstacles to expansion, these
provider-driven schemes were redesigned from 2006 on, aiming at a switch towards a more
community-based organisational culture. The evaluation concluded that such switch had not
been made. Service providers still regarded and treated the schemes as their property.
Besides, scheme managers still lacked necessarily knowledge and skillscxv.
Institutional design became manifest from 2005 on, when the Ugandan government
launched a programme to promote CHI. Currently the government is designing a National
Health Insurance Scheme (NHIS), which contemplates the integration of existing schemescxvi
and for which in 2009 a National Health Insurance Bill was drafted. Unlike the homonymous
56 Faith-based hospitals make up half of all district hospitals in rural Uganda. 57 This is case of the ‘Save for Health’ scheme in Luwero, central Uganda.
40
construct in Ghana, the Ugandan NHIS envisages a fragmented strategy, with separate
insurance arrangements for different population groups and – a particularity of the Ugandan
approach – introduction of these arrangements in a phased manner. According to the draft in
circulation, the future NHIS’ first building block would be the enrolment of public servants.
Researchers have criticised this scenario for possibly resulting in too low initial coverage,
little cross subsidisation and limited financial protectioncxvii. It should however be noticed
that the design of the NHIS – although at an advanced stage – is not yet carved in stone. Also
according to the present draft of the NHIS, CHI would be applied in both the private and the
government sector. The latter option unveils the somewhat ambiguous position of Ugandan
policymakers, who seem to have great difficulty to openly admit the limited success of user
fee abolition in the country.
In recognition of the need for capacity building at CHI management levelcxv – and an
expected even bigger need when the schemes will be integrated in the future NHIS – the
Uganda Martyrs University (UMU)58 will launch an advanced diploma course in health
insurance management in 2011.
Data on empowerment and transformation through Ugandan CHI are lacking.
Rwanda
Gains in terms of financial coverage at country level seem to be in contradiction with the
impressive gains in CHI population coverage: out-of-pocket expenditure as a proportion of
58 See http://www.fiuc.org/umu/
41
private health expenditure rose from 40.7 to 44.4% between 2000 and 2007xlii,59 – while CHI
population coverage had surpassed 75% in 2007cxviii. This paradox – though significant –
should not distract from the exceptional progress that Rwanda made in just one decade, or
from the lessons that Rwanda has to offer.
Mutual organisations in Rwanda date back to pre-independence years; community health
insurance schemes are a more recent phenomenon60. In the early 1990s, a number of
provider-driven schemes still subsisted or were reactivated after the reintroduction of user
fees in 199661, some of them more successful than otherscxix. In 1999, a state-driven approach
– part of the national reconstruction effort following the 1994 genocide – took over. At that
time population coverage by CHI was as low as 1.2% and health centres had an utilisation
rate of 0.28 new cases/inhabitant/year on averagecxviii. With technical and financial
assistance of Abt Associates and USAID, the government piloted 54 health centre-based
Mutuelles de Santé in three districts from July 1999 oncxx. After one year of piloting, the results
were considered promising: 8% of the population in the targeted areas had enrolled; health
service utilisation was six times higher for members than for non-members; members
contributed per capita five times more to health services than non-members while spending
less out-of-pocket per episode of illness than non-memberscxx,cxxi.
Following the 1999-2001 pilot phase, the Mutuelles approach was progressively spread out. In
2003, 102 schemes covered 7% of the national population. In 2004, 226 schemes covered 27%;
in 2005, 354 schemes covered 44% - nearly four million peoplecxviii, , , ,cxxii cxxiii cxxiv cxxv. Utilisation
59 The share of non-mandatory prepayment plans in private health expenditure is no longer a relevant indicator for the impact of CHI in Rwanda, since CHI in principle became mandatory in 2007. This share rose from 0.9 to 10.2% between 2000 and 200742. 60 Pioneers in Rwanda were the Muvandimwe scheme in Kibungo (1966) and the Umubano mubantu scheme in Butare (1975)122,123. 61 In 1998 – before the government took the lead – a total of six schemes were counted123.
42
rates of the health centres increased concomitantly, reaching 0.42 new cases/inhabitant/year
in 2004 and possibly 0.65 in 2005cxviii, , ,62 cxxvi 63.
Alongside the striking gains in population coverage during the 2002-2005 expansion phase, a
series of challenges remained. To name a few: affiliation was still voluntary, which implied a
risk of adverse selection despite the large numbers; service coverage was still patchy and
exclusive of big risks at referral level; overall financial stability and peripheral management
capacity were weak. The Rwandan government recognised these challengescxxiii, adapted
design and practice64, and proceeded. In a third – still ongoing – phase of consolidation,
Rwanda reported 73% population coverage in 200665, 75% in 2007, and 85% in
2008cxxiv,cxxv,cxxvii. Utilisation rates of curative services66 reached 0.61 new
cases/inhabitant/year in 2006, 0.72 in 2007, and 0.86 in 2008cxxv.
A closer look at data provided by the 2005-2006 Rwandan Integrated Living Conditions
Survey (EICV2, Enquête Intégrale sur les Conditions de Vie des ménagescxxviii ,) reveals insight on
the Mutuelles’ financial coverage. Out-of-pocket payment for health67 of CHI members was
half that of non-members, both in absolute terms and as proportion of capacity to
paycxxix, ,cxxx 68.
62 Several records mention low service utilisation (0.28 new cases/inhabitant/year in 1998 at health centre level) as one of the drivers for the Rwandan CHI initiative118,121,122,127,128. The Rwandan government explicitly names three objectives and one expected outcome of its CHI policy: “1) to improve financial access to health care, 2) to improve the financial situation of health facilities, and 3) to improve the overall health status of the population. Mutual health insurance should facilitate the utilization of services by the population”122,111. 63 Musango and colleagues reported an utilisation rate of curative services of 0.65 in 2005, based on figures from the 2007 Year Report of the Ministry of Health118. The 2009 Statistical Yearbook refers to 0.47 in 2005, based on figures from the 2008 Year Reports of the Ministry and the Mutuelles125. 64 More on this in the following section on institutional design and organisational practice. 65 Within the sample of the 2005-2006 Integrated Living Conditions Survey (EICV2), CHI population coverage was 36%128,130. 66 Combined for health centres and referral hospitals, with the portion at hospital level being less than 5% of the total125. 67 Including the cost of consultation, hospitalization, drugs, tests and transport, but not the insurance premiums paid. The survey used a recall period of two weeks128. 68 The previous EICV (EICV1, 2001-2002) had no separate data for CHI members and non-members. Deduction allows however to assume an increase in financial protection between 2000 and 2006: in real terms, health expenditure for the total population had decreased; health expenditure for non-insured had remained the same128.
43
Due to the Rwandan government’s strong involvement in CHI since 1999, institutional
design and organisational practice are manifestly intertwined. Furthermore, Rwanda’s CHI
policy is today part of a bigger aggregate including performance-based payments (PBF) and
enforced donor coordinationcxxvii. Analysis of design and impact of these interactions is of
major interest.
Already in the pilot phase (1999-2000) important choices were made: linkage of each CHI
scheme to a health centre and introduction of capitation payment; and affiliation to CHI per
family and contributions linked to presumed capacity to pay69. These measures had both
positive and negative impact. While capitation payment and low personal contributions
helped to control cost for members and schemes, health centres were neither prepared nor
equipped to the consequences. Besides, limited benefits packages and insufficient
involvement of local authorities were identified as serious weaknesses. While the overall
impact of the pilot project was positive enough to decide for expansion, design and
implementation improvements were deemed necessarycxviii,cxxi.
The Rwandan government started the phase of expansion (2002-2005) making local
authorities responsible for CHI development70 and concluded it preparing a specific legal
framework for health insurance. Operational roles and responsibilities for CHI at different
levels were clearly laid out. Note was taken of the remaining financial barriers for the
poorest of the poor. A system for identification of the extreme poor was established as well
as prepayment of premiums through micro-credits at the community level. Note was taken
also of the need of subsidies to strengthen the CHI schemes and to widen their benefits
69 Estimated as a proportion of yearly income; not calculated on the basis of declared willingness to pay. In the pilot phase, the personal contribution of a family of seven corresponded on average with 8% of the family’s yearly income, which was deemed fair118,121. 70 “Les préfets des provinces, les maires des villes et des districts administratifs ont reçu des directives de leur ministère de procéder à la mise en place des mutuelles de santé dans leurs zones respectives dans des délais aussi courts que possible. (…) La création des mutuelles de santé sera un des points de leur évaluation future.” Minister of local administration and social affairs, 2003, cited by Musango and colleagues118.
44
package. Gradually, a comprehensive action plan for the coming years was drafted and
budgetedcxxiii.
Government made major design adjustments entering the phase of consolidation. In 2006, it
started subsidising part of the premium of the extreme poor, and succeeded in channelling
Global Fund money through CHIcxviii,cxxiv,cxxxi. It thereby intelligently bypassed a trade-off
between extension of coverage and maximising revenues for the health system through
CHIcxxxii. While the Rwandan CHI initiative had always been part and parcel of the national
poverty reduction strategy, the latter was now reinforced with an agreement to integrate all
donor funds within one fiscal framework. And hand in hand with CHI at the demand side,
performance-based payment was introduced at the supply sidecxxvii. These measures have
certainly contributed to the gains in coverage and access over the last years. They have also
made evaluation of CHI per se more complex71. Finally, following a 2006 ministerial decree,
in 2008 a specific legal framework was established, making affiliation to health insurance in
principle mandatory for nationals and residents alikecxviii.
Clearly, this comprehensive and diverse package of measures was the start rather than the
conclusion of a process. Recent research highlights persistent and newly discovered
challenges. The contributions made to the extreme poor are insufficient to overcome their
difficulty to adherecxxxiii. Despite the efforts made, quality of care on offer is far from optimal:
a 2007 service provision assessment found a full package of basic services available in 44%
only of all health facilities, and low staff adherence to treatment standardscxxvi. Integration
with health insurance mechanisms for other population groups – especially with RAMA (La
Rwandaise d’assurance maladie) serving the civil servants – is desired but not yet realisedcxviii.
71 This being an academic, more than a practical issue.
45
Financial sustainability remains a substantial, and monitoring of implementation remains a
huge challengecxxiv.
The presence and nature of a specific dynamic of empowerment and transformation would
be a proper subject of monitoring and research, which has barely received attention so far.
One exception is to be found in the work of Kalk and colleagues on the Rwandan Mutuelles
and health system. They describe how CHI scheme management by civil servants is a
possible barrier to genuine community participationcxxxi,cxxxiii.
Burundi
The Burundian government’s effort to spread some form of community health insurance
preceded the Rwandan state-driven approach by a decade and a half, but was far less
successful. Of Burundi’s Carte d’Assurance Maladie little remains but lessons from a remote
past.
A prolonged political and humanitarian crisis left its mark on the country, and few
international actors lent a hand. Today, the poorest Central African state is still in fragile
conditions. Two local CHI initiatives are active; a few others are starting out.
Gains in terms of financial and population coverage are insignificant or at best marginal at
country level: out-of-pocket expenditure as a proportion of private health expenditure
dropped from 71.3 to 60.5% between 2000 and 2007, while the share of prepayment plans
(including CHI, not including mandatory health insurance) in private health expenditure
actually dropped from 0.4 to 0.2% over the same periodxlii.
46
In 1984, the Burundian government installed the CAM scheme (Carte d’Assurance Maladie).
This was a national programme, with revenue collection and management at community
level. In the initial setup, purchase of a card (CAM) for a fixed annual premium of less than
two dollar in principle entitled a household (up to two adults, plus all children under 18) to
one year of inpatient and outpatient care – without copayment and including drugs – in all
public facilities, at both primary care and referral level. Despite its apparent attractiveness,
the CAM never reached more than 10-25% of the population.
In 1996, a differential annual premium (three categories72) and a 20% copayment were
introduced. All along successive crises from 1993 on, the CAM waned into near oblivion, for
potential users and service providers alike. Failing protection went to extremes. In 2006,
human rights activists reported detention of insolvent patients as routine practice in
Burundian hospitals. Of those detained with information of insurance coverage available,
59% had in vain presented a CAM cardcxxxiv, , ,cxxxv cxxxvi cxxxvii.
Burundi’s unsuccessful CAM experience provides important lessons in terms of institutional
design and organisational practice. As early as in 1993, Baza and colleagues described the
CAM as both a success and a failure: a success because as many as 95% of users of public
health centres had a card, and a failure because as few as 10% of the population procured a
card. They further inquired why uptake of the CAM – all in all an insurance arrangement
that promised extensive benefits – was so low. Above all, people perceived the CAM as no
value for money, as drugs were out of stock more often than not. Besides, no supply-side
measures complemented the top-down design of the scheme. In fact, no link was foreseen
72 The 1996 amendment restricted the CAM to the self-employed and set differential premiums according to strata within this group: 500 Fbu (Burundian francs) a year for farmers, pastors and fishermen; 1,500 Fbu a year for artisans, retailers and shopkeepers not registered with the tax services; and 3,000 Fbu a year for artisans, retailers, shopkeepers and other self-employed registered with the tax services140 (respectively 2.25, 4.5 and 13.5 US$ a year at 1996 exchange rates; the Fbu has since devaluated more than 500%).
47
between payment for the CAM and financing of the health services. People regarded the
CAM both as a gift and as a tax, albeit a useless onecxxxiv. Arhin, in her 1994 study, came to
similar conclusionscxxxv. Both authors pled for improvements in quality of care, among other
means by linking CAM revenues to supply-side investmentscxxxiv,cxxxv, but no action was
taken.
By 1999, an in-depth World Bank poverty note described Burundi as on its way “to a full-
scale humanitarian emergency”. Regarding social protection in health, it described how lack
of public resources73 and escalating needs had ruptured the CAM. By the time, all public
health centres were charging for drugs despite the CAM74. In its conclusion, the document
pled for urgent action in social protection, to be designed and implemented with community
involvement and ownershipcxxxviii. Such did not happen.
From 2006 on, first attempts of recovery were noted at country level. At that time – according
to data from a Core Welfare Indicators Questionnaire (QUIBB, Questionnaire des indicateurs de
base du bien-être)75 – one out of three Burundians did not seek healthcare when in need, of
which eight out of ten because they could not afford it. Burundi engaged in a poverty
reduction strategy, launched its 2006-2010 National Health Sector Development Plan (PNDS,
Plan National de Développement Sanitaire) and abolished user fees for under-five and maternal
healthcare. The 2006-2010 development plan mentions – among a total of 68 strategic actions
– promotion of CHI to improve financial access, in particular for the extreme poorcxxxix,76.
73 The 1999 poverty note also attributed part of the health systems crisis to the withdrawal of donors138. This withdrawal would continue in the next decade. 74 Whereas some nonprofit providers (who never subscribed to the CAM) temporarily reduced or eliminated cost-recovery measures138. 75 Cited in a World Bank project appraisal document for health sector development support; see http://www-wds.worldbank.org/external/default/WDSContentServer/WDSP/IB/2009/05/21/000333038_20090521015520/Rendered/PDF/472600PAD0P101101Official0Use0Only1.pdf 76 This portion of the population is estimated at above 80%, since over a decade. Gross national income was as low 380 US$ per capita in 2008. See http://siteresources.worldbank.org/INTWDR2010/Resources/5287678-1226014527953/Statistical-Annex.pdf
48
A 2009 inventory of social protection documented two active CHI networks in Burundi: the
provider-driven MSAG (Mutuelles de Santé de l’Archidiocèse Gitega) – integrating 14 schemes,
jointly covering 18,074 people – and the union-based MUSCABU (Mutuelles de Santé des
caféiculteurs du Burundi) – integrating 15 schemes, jointly covering 39,175 people. Others are
starting or preparing to do so. All receive assistance from Belgian nongovernmental
development partnerscxl.
Currently, Burundi sets out for a health sector reform. Elements included are performance-
based financing and donor coordination within a sector-wide approach. Among the few
donors active in the Burundian health sector is the Belgian Technical Cooperation
(BTC/CTB). Among other projects, BTC/CTB will focus in 2011 on social health protection,
aiming at coordinating the already existing CHI efforts and at revitalising the Carte
d’Assurance Maladie.
Not unsurprisingly, the largely disappointing early CAM experience still offers positive
evidence on empowerment. As stated by Arjin in 1994, women reported that the CAM
empowered them to decide the need for – and the timing of – healthcare consumption. This
finding bears a striking resemblance to what Soors and Criel recorded in Nongon (Mali)
more than a decade laterlxxii,77, and highlights the potential of CHI as a social
investmentxviii,cxxxv.
77 See also page 21.
49
Democratic Republic of Congo
The Democratic Republic of Congo (hereafter called DR Congo) is today possibly one of the
most challenging contexts for any kind of development, including that of CHI. Yet new CHI
initiatives in the DR Congo are common. A 2004 survey of 28 Congolese schemes highlighted
evidence of a countrywide renewal in CHI activities since 2000, and to a great variety of CHI
modelscxli.
Little can be said in terms of financial and population coverage at country level: out-of-
pocket expenditure as a proportion of private health expenditure dropped from 97.0 to 51.7%
between 2000 and 2007, apparently without relation with the share of prepayment plans
(including CHI, not mandatory heath insurance), which was reported 0% in 2000 and 2007xlii.
While the DR Congo witnesses a boom of CHI activity, scheme-specific data are rare. One
notable exception is that of the Bwamanda scheme. This African CHI pioneer – active since
1986 as part of a broader development projectxxxiii,78 – survived political instability, economic
decline and war, and has information on population and service coverage to offer.
Within three years, the scheme’s coverage had increased from (an already impressive) 28%
to 60-65% of its target population, where it stabilised for a decadecxlii. Subscription declined
during wartime, but began to increase again from 2003, peaking at 64% (114,465
beneficiaries) in 2004 before settling around 55% until 2008cxliii.
78 The CDI (Centre de Développement Intégral) Bwamanda is a Congolese NGO – supported by a Belgian counterpart and several development partners – with agricultural, educational, infrastructural, ecological and medical activities, in the northwestern part of the Equator province, since 1969.
50
Service coverage of the Bwamanda scheme is limited to hospital care, contingent on referral
from a health centre79. The scheme improved significantly hospital utilisation for its
memberscxlii, and more so in case of priority interventions, to the extent that scheme
membership was to overcome geographical barriers for patients considered in high
needcxliv,80.
Elements of organisational practice – and wider contextual elements – proved substantial
for the success of the scheme in terms of coverage and access. The Bwamanda scheme was
launched in a setting where the quality of the care supplied was of a relatively high
standard, where people basically trusted the management of the scheme and where the
management team had the freedom, willingness and skills to test change81. The set-up of the
scheme was exhaustively negotiated with the community and comprised collection of
membership fees coinciding with annual crop sales, community-rated flat membership fees,
coverage contingent on referral, low co-payments and a cashless system at the time and
place of utilisation for the proportion of cost covered82, and scheme management by the
district health teamlxii,cxlii. It can be argued that the fit of organisational practice to the
community’s expectations has been key for the scheme’s exceptional success, survival during
and revival after the war.
79 The scheme’s environment being a two-tier health district with 23 health centres and one referral hospital127. 80 Criel and colleagues’ in-depth study on hospital utilization at Bwamanda considered C-sections and strangulated hernias as priority interventions144. 81 The Bwamanda health district was already relatively successful before the insertion of the CHI scheme (0.6 new cases/inhabitant/year utilisation rate of outpatient care and 84% ANC coverage in the health centres; 30/1,000 admission rate in the hospital). Paradoxically, decay of the (at the time) Zairese state and the resulting isolation of the district was a main driver in the search of self-reliance, including the CHI scheme142. 82 The community had expressed a preference for no co-payments at all. Eventually, a consensus was reached on no co-payment for maternity services, and a 20% co-payment for all others. The 100 and 80% coverage by the scheme was directly transferred from scheme to hospital upon utilisation of care, thereby avoiding the hassle of retrospective reimbursement for the users142.
51
From 2008 on, new conflicts and resulting crisis in the region started taking their toll.
Provision of drugs became problematiccxlv. The scheme’s population coverage dropped to
26% (49,858 beneficiaries) in 2009cxliii and is slowly recoveringcxlv.
All over the DR Congo, new schemes keep popping up in recent years. While part of this
occurrence can be seen as a reaction to a failing health system, institutional design is slowly
lending a hand: the Congolese government today encourages CHI development, as
expressed in the 2009-2011 action plan of its National Programme for the Promotion of
Community Health Insurance (PNPMS, Programme National de Promotion des Mutuelles de
Santé)cxlvi,83. What the long-term effect of these many new schemes will be remains to be
seen.
Understandably, effects on empowerment and transformation are yet undocumented.
83 See also the PNPMS’ 2009-2010 logframe: Cadre logique du Programme National de Promotion des Mutuelles de Santé. http://www.coopami.org/fr/countries/countries_partners/drc/projects/2009/pdf/20092010programpnpms.pdf
52
Community Health Insurance in Asia
Unlike in Africa, the earliest CHI initiatives in Asia were rooted in deliberately political
processes. In China, the first Medical Cooperatives saw the light of day in a few communist-
controlled rural areas as far back as the 1940s. Once an isolated innovation, these schemes
eventually led to the nationwide implementation of the Rural Cooperative Medical System
(RCMS) in the 1960s. By the 1970s, RCMS covered 90% of China’s rural population.
However, the RCMS collapsed following the market-oriented reforms of the early 1980s. A
new RMCS was created in 2003 and is in expansion.
In the Indian subcontinent, the Students Health Home (SHH) was the first CHI scheme to be
recorded, set up by a communist movement in West Bengal back in 1952cxlvii. It was not until
the late 1990s however that a crossover between the micro-finance movement (which
originated mainly as micro-credit, but now encompasses a variety of products including
micro-credit, micro-savings and micro-insurance) and CHI initiatives led to a spurt of CHI
schemes, mainly in India and Bangladesh.
As is the case with CHI in all contexts, different speeds of implementation, variations in the
mode of implementation and heterogeneous achievements can be observed. With regard to
country patterns, roughly three groups can be distinguished – of which we discuss two
examples each: early CHI developers, where CHI today plays again (China, page 62
onwards) or still (India, page 71 onwards, with brief mention of Nepal and Bangladesh) a
prominent role; late and possibly innovative developers (like Cambodia and Lao People's
Democratic Republic, pages 82 and 86 onwards); and early developers where CHI actually
only plays a minor role (like the Philippines and Indonesia, pages 88-89 and 90-91).
53
China
The early Medical Cooperatives in the Shanxi, Gansu and Ningxia provinces were
established as a mechanism to help defray cost of medical treatment and drugs. A first set up
as mutual prepayment funds, they subsisted on the peasants’ voluntary contributions in the
form of both cash and in kind, as well as initial drug stocks provided by the ruling
communist local governments. These initiatives proliferated and gained financial strength
during the 1950s, when the communist state organised the agricultural workers into farmer
cooperatives and consequently was able to introduce welfare funds at community level. As
an integral part of the collective system for agricultural production and social services, the
Rural Cooperative Medical System (RCMS) became a nationwide structure of prepayment
schemes for healthcare financing during the 1960s. Most villages funded their Cooperative
Medical Scheme from three separate sources: household health insurance premiums, a
collective welfare fund and state subsidies. Depending on the plan’s benefit structure and
the village’s economic status, the household premium was usually fixed at between 0.5 and
2% of a peasant’s family annual income. The welfare fund was a state-defined portion of the
village’s collective income from agricultural production. Subsidies from upper-level tiers of
governments were typically earmarked to compensate health workers and purchase medical
equipment. In 1965, the state explicitly encouraged the entire rural sector to adopt the
Cooperative Medical Scheme (CMS) as the mode of financing and organizing healthcare
services. The resulting community financing and organization model is believed by many to
have contributed significantly to the achievements of the Chinese primary care of that
eracxlviii. Between 1949 and 1973, the infant mortality rate was reduced from about 200 per
1,000 to 47 per 1,000 live births and life expectancy increased from 35 to about 65 years. From
54
the late 1960s until 1979 – when the process of collectivization began to be reversed – the
RCMS covered 90% of China’s rural residentscxlix, ,cl 84.
Due to market-oriented reforms both the communal administrative structure that employed
the health workers and the collective welfare funds – once counting for 30 to 90% of the
schemes’ funding – disappeared. By 1984, RCMS population coverage had dropped to less
than 5%. Between 1981 and 1993, the contribution made by the RCMS to national health
expenditure fell from 20 to 2%. Besides, poor management practices and corruption had
eroded the population’s trust in the once successful schemescl,cli. Attempts to re-establish the
RCMS based on voluntary affiliation and without substantial co-funding by central
government had limited success. By the end of the century, 90% of China’s rural residents
were uninsuredclii,85.
Still, the Rural Cooperative Medical System never disappeared entirely from the political
agenda. In 2002, the Asian Development Bank made an appeal for its reinstatement, at least
in China’s middle-income regions, a plea contingent on renewed and committed government
supportcxlix. In 2003, China created a new RCMS – also known since as NCMS, New
Cooperative Medical System – projected to cover the whole of rural China by 2008cliii,86. The
NCMS is based on voluntary affiliation (unlike the RCMS before the reforms of the 1980s), is
co-funded by the local and central government (as had not been the case since the reforms of
the 1980s), and is managed on county level. Central government declared coverage of
84 In the pre-reform period almost all Chinese citizens were covered by some form of health insurance: besides RCMS covering most of the rural population, the urban population benefited either from the Labour Insurance Scheme (LIS, for state-owned enterprise workers) or from the Government Insurance Scheme (GIS, for civil servants)157. 85 Insurance coverage (LIS and GIS) of the urban population declined to a less extent, arriving at around 50% in 1998157. 86 At inception, this projection was made for 2010. Rapid expansion of insurance coverage (86% of the rural population in 2007139) led to reformulation of this deadline153.
55
catastrophic illnesses87 a key aim of the NCMS; but the actual decisions on benefit package
and provider payment method are left to the local management structurexxii, , ,cliv clv clvi.
The creation of the NCMS for the rural population was part of a broader health insurance
and social assistance reform in Chinaclvii, , ,clviii clix clx, including a Basic Medical Insurance (BMI)
initially aiming at covering all urban workers, since 199888; a Medical Financial Assistance
(MFA) safety-net programme for the urban and rural extreme poor, since 2003 and closely
linked to the NCMSclxi, , , ,clxii clxiii clxiv 89; and an Urban Resident Basic Medical Insurance
(URBMI) for urban residents not covered by the BMI, since 2007clxv,90.
Gains in terms of financial coverage are modest at country level: the share of prepayment
plans (including CHI, not mandatory health insurance) in private health expenditure rose
from 1.0 to 7.1% between 2000 and 2007, while out-of-pocket expenditure as a proportion of
private health expenditure dropped from 97.3 to 92.0% over the same periodxlii.
This indication of an elusive impact on financial protection contrasts remarkably with the
reported expansion in population coverage: according to official data, the NCMS covered
73% of the targeted rural population in 2004clvii, 86% in 2007xxii and 95% in 2008clv. Yet
insurance coverage not automatically leading to financial protection is no new phenomenon
in China. As early as 1996, Bogg and colleagues had reported increased out-of-pocket
87 Called dabing, literally severe illnesses155. Commonly defined as acute illnesses associated with inpatient care156. 88 The BMI – also known as UEBMI (Urban Employee Basic Medical Insurance – was piloted from 1995 on and spread out from 1998 on. Classified by most scholars as a social health insurance scheme and based on mandatory affiliation, it aimed at covering all urban formal-sector and government workers, but not their dependents. Coverage of the urban population was still below 40% in 2003158. The large rural-to-urban migrant population – and informal-sector workers in general – is not eligible for BMI165. 89 Strictly speaking, the MFA is no insurance, but a safety net. Because however it is closely linked to – and dependent upon – the NCMS, it makes sense to pay closer attention to MFA in this overview. Indeed, MFA was designed as complementary to NCMS (being an entry point to NCMS by paying the premium for those to poor to contribute, or acting as ‘second assistance’ by transferring cash after reimbursement by NCMS), and its implementation and effectiveness depend on the performance of NCMS (and of present social assistance schemes like Wubao [‘Five Guarantee Programme’], Dibao [‘Minimum Income Guarantee Scheme’] and Tekun [‘Assistance for the Extremely Poor Households’])162. 90 As NCMS and unlike (UE)BMI, URBMI – which according to the official guidelines targets “primary and secondary school students who are not covered by the urban employee medical insurance system (…), young children, and other unemployed urban residents” – is based on voluntary affiliation165.
56
spending in a county after introduction of one of the predecessors of NCMScl. Based on
survey data reaching from 1991 to 2003, Wagstaff and Lindelow found that health insurance
in the pre-NCMS era not only increased out-of-pocket but also the risk of high and
catastrophic spendingclviii. Wagstaff and colleagues argued that insurance had led to these
unexpected effects by allowing healthcare providers to deliver more costly care and by
pushing healthcare users towards higher level (and more costly) providersclix. There is today
no evidence that the NCMS would have reversed these perverse incentives, and at most
evidence for a marginal positive impact on financial protection.
A 2005 study in rural Shandong provides evidence that NCMS adversely influences
prescribing behaviour of village doctorsclxvi. Sun and colleagues’ impact evaluation based on
data from the same county under study found that NCMS reduced the incidence of
catastrophic health expenditure (CHE) by no more than 8.2% in 2004 (from 8.98 to 8.25% of
households) and the severity of CHE by 18.7% (from 8.95 to 7.28 times a household’s
disposable income) in 200491. As the authors recognized, a CHE incidence of 8.25% is yet
high, and still spending a 7-fold of the disposable income is a disasterclxvii. A simulation
based on 2006 survey data assessed the effectiveness of NCMS in financial protection. The
authors computed a 3.5-3.9% reduction in poverty headcount, and concluded that the NCMS
offered only limited financial protectionclxviii. Yet another simulation compared out-of-pocket
spending before and after the introduction of NCMS, and concluded that NCMS had not
reduced out-of-pocket expenses per outpatient visit or per inpatient episodeclxix. Wagstaff
and colleagues repeated this conclusion in a comprehensive review of recent studies of
China’s healthcare reformclix. A more focused review – on the NCMS only, by You and
Kobayashi – likewise found no effect of the NCMS on out-of-pocket expenditureclxx.
91 The authors used the term catastrophic health payment as identical to catastrophic health expenditure and applied five thresholds (spending more than 10, 20, 30, 40, 50 or 60% of disposable income on healthcare). For this simulation, they applied the 40% threshold167.
57
Examining data from the nationwide 2000, 2004 and 2006 China Health and Nutrition
Surveys, Lei and Lin found no effect of the NCMS on out-of-pocket expenditure eithercliii.
Examining data from more recent national surveys in a 2005-2008 panel study, Yi, Zhang and
colleagues found that reimbursement by NCMS has an increasing tendency of favouring
low-level expenditures, and fails to cover its objective of minimising the financial
implications of catastrophic illnessesclv,clxxi. Examining data from a 2008 survey in Western
and Central China, Shi and colleagues observed that NCMS reduced the incidence of
catastrophic health expenditure (CHE) by no more than 9.9% and the intensity of CHE by
16.9%clxiii,92. A 2010 analysis of 2004 and 2007 data, by Babiarz and colleagues, arrives at the
more positive picture of a 19% reduction in out-of-pocket payment but gives no conclusive
evidence on catastrophic health expenditureclxxii,93. As stated in an accompanying editorial,
the results at which Babiarz and colleagues arrive should be interpreted with caution, in
view of their rather thin evidence base and the contradictory findings from other studiesclxxiii.
Several articles attribute the NCMS’ past and current failure of providing its essential
function of financial protection (at least partly) to insufficient fundingclv,clxiii,clxvii,clxviii,clxxi. One
article sees improved performance over time, attributable to a variety of policy modifications
including increased governmental fundingclix. In 2006, the Chinese government committed to
increasing healthcare funding by 1-1.5% of its gross domestic product – which would
represent a triple of government spending on health – subsidizing both the supply and
demand side, including increased subsidies for the NCMSxxii. There is today a growing
consensus that increasing funding alone will not be sufficient, and could even be
92 The 2008 figures on reduction of incidence of CHE provided by Shi and colleagues (9.9%) are similar to the 2004 figures on the same provided by Sun and colleagues (8.2%), and comparable: both groups of authors used a 40% threshold to esteem incidence of CHE. The 2008 figures on intensity of CHE should not be compared with the 2004 figures on severity of CHE: Shi and colleagues utilised intensity of CHE as the degree by which health expenditure exceeds the 40% threshold, whereas Sun and colleagues utilised severity of CHE as the relation of health expenditure to disposable income at the 40% threshold163,167. 93 Instead of using CHE as indicator, the authors calculated the probability of financing healthcare through asset sales or borrowing, for which they observed a 45% reduction173.
58
counterproductive as far as financial protection is concerned. Both Yip and Hsiaoxxii and You
and Kobayashiclxx argue that more money will most likely end up as income and profits for
providers, unless cost inflation and inefficiencies will be carefully tackled. This hypothesis
highlights the importance of institutional design and organisational practice, which we will
deal with more in detail in a following section.
Financial coverage by the Medical Financial Assistance (MFA) displays a yet bleaker picture.
With its implementation decentralised to county level94, as is the case with the NCMS to
which it is complementary, MFA in most counties covered only a small fraction of actual
medical costs and faced sustainability problems already in its development stageclxi. A 2006
study in four counties affirmed insufficient financial protection of MFA, due to exclusive
retrospective reimbursement95 and lack of fundingclxii. A 2008 survey in Western and Central
China documented only one case of prevention of impoverishment – and no protection
against catastrophic health payment – by MFA after three years of implementation. The
authors of the study advocate a range of design improvements, contingent on increased
fundingclxiii. Another study found a positive impact on utilisation of care – but not on
impoverishment – in a pilot setting with an extended benefit package. The authors of this
study advocate a more substantial extension of the MFA benefit package, and regard the
government’s commitment to increased healthcare funding as an opportunity to explore this
possibilityclxiv.
Impact by the NCMS (and MFA) on utilisation of care – and the interaction of utilisation of
care with financial coverage – is worth a closer look. There is indeed ample evidence of a
positive overall impact of both NCMS and MFA on utilisation of care, which might suggest a
94 This resulted in a marked variation across counties in all three dimensions of coverage, due to considerable discretion over financing, eligibility, the range of services covered, the proportion of costs covered and payment methods161. 95 A non-quantified fraction of the enrolled poor could not afford to pay for medical services in advance, and could thus not benefit from MFA in these four counties162.
59
reduction in unmet needs. Wagstaff and colleagues, in their simulation exercise before and
after the introduction of NCMS, found that outpatient visits had increased by 52%, and
inpatient stays by 47%clxix. Hao and colleagues examined data from a 2004 survey and found
increased utilisation after introduction of MFA, especially for inpatient care and more so
when the benefit package was extendedclxiv. Zhang and colleagues, in their panel study on
data from 2005 and 2008 surveys, found that outpatient care in case of need had increased
from 90 to 95%, and the share of those who used inpatient services from 7 to 10%clxxi. Lei and
Lin examined data from 2000, 2004 and 2006 survey and came up with mixed findings:
uptake of preventive services had increased, utilisation of curative services had not. They
ascribed the latter among other factors to the NCMS’ prohibitive deductibles and
insufficiently funded saving accountscliii. In their 2009 review, Wagstaff and colleagues
confirmed a positive impact on service utilisation, but warned that without tight measures to
alter the persistent and perverse incentives for healthcare providers, broader and deeper
insurance coverage would continue to translate in more resource-intensive care, and in
possibly less instead of more financial protectionclix.
Accumulating evidence from the interplay of reproductive health and current Chinese
insurance efforts provides a similar warning. Back in 1995, Bogg and colleagues conducted
an extensive study on the determinants of uptake of maternal care services in Central China.
One of their findings was that cost recovery was a strong barrier for the use of antenatal care
services (ANC). While only few women at the time benefited from existing maternal
insurance schemes96, those that did had a 1.6 times higher possibility to use ANC than those
that had no insurance coverageclxxiv. Another finding was a strong association between home
delivery and having no insurance coverage. Fifteen years later and with the NCMS spread
96 Between 0 and 15% of eligible women in five out of six counties under study; 43% in the sixth174.
60
out over rural China, the situation looks different. Comparing a county where NCMS
covered ANC with two counties where ANC was not included in the NCMS benefit package,
Long and colleagues found no significant difference in ANC uptake between covered and
uncovered. Besides, of the group with ANC coverage, less than half of the women were
aware that ANC was covered and only 20% claimed reimbursement. Out-of-pocket
expenditure for ANC remained at 8% of an average woman’s annual income, and was as
high as 26% of the annual income of a woman in the poorer third of the population.
Regardless of ANC coverage, providers under NCMS did not abide with national ANC
guidelines: they rarely prescribed all recommended tests (and less so in the county where
ANC was covered), yet exceeded in 90% of the cases the recommended number of
ultrasounds (in all three counties)clxxv. Bogg and colleagues in 2010 report an alarming
increase in C-section rates after introduction of the NCMS (2004-2007)97. While the authors
do not claim that NCMS funding was the sole or primary cause, they do suspect that the
NCMS provider payment mechanisms in combination with the dependence of doctors’
income on hospital revenues are the culpritsclxxvi.
In terms of organisational practice and even more so of institutional design, China’s early
21st century post-reform reforms are certainly one of the richest sources for health policy and
social security researchers today. They are also one of the most complex, and difficult to
interpret. From a theoretical policy view, China’s course to universal coverage can be
described as the biggest ongoing experiment of a fragmented approachxlviii,lxxi, with CHI-like
initiatives (NCMS, URBMI) being part of an aggregate including social health insurance
(UEBMI) and social assistance in health (MFA). From an operational research perspective,
China’s myriad of designs and implementation modes within the established NCMS and
97 An increase of between 36 and 131% in four of the five counties under study; a status quo in the sixth where C-section rate already was at an unacceptable 60%176.
61
MFA schemes is innovative and provides a wealth of policy lessonsclxix, , ,98 clxxvii clxxviii. From the
angle of many low-income country planners and policymakers, China’s impressive and fast
achievements in insurance population coverage are desirable, if not exemplary. Yet the non-
translation of population coverage in financial protection is a challenge for policy theorists,
and more so for Chinese policymakersxxii,clix,clxviii,clxix. And eventually, the mere feasibility of
reaching universal coverage through a fragmented approach is at stake.
The specific dynamic of empowerment seems not to have captured the interest of scholars
of China’s healthcare and social health protection reforms. While a multitude of authors have
described how China’s social and economic transformations led to a felt need for social
protection in health including community health insurance99 and many have pointed to a
reinforcement of existing patient-provider asymmetries after introduction of insurance
schemesclxxix,100, no research specifically addresses the question if the schemes have led to
any positive transformation and patient empowerment. One group of researchers – in an
overview article on regulation of the healthcare markets in China and India – observes that
in China “the role of civil society organizations (…) is very underdeveloped, and the degree
to which schemes can be influenced to reflect the interests of their beneficiaries is
uncertain”clxxx. With reference to the NCMS, a recent BMJ editorial brings up an underlying
98 Wagstaff and colleagues (2009) described the piloting of NCMS as a policy of “letting a thousand flowers boom”, and saw this characteristic as a deliberate design feature, to ensure that lessons could be learnt from local experimentation169. Studying the emergence of the MFA programme, Xu and colleagues (2008) noted that its implementation process typifies the Chinese approach to deliver social protection programmes: experiment first and adjust policies later through learning-by-doing161. Taking an urban social assistance programme (known as the Minimum Living Standard Guarantee System) as an example, Leung (2006) described how local governments were encouraged to experiment with different models, and central government would evaluate and attempt to unify diversified practices by promoting one of the more successful models177. In fact, this particular way of gradual reform – cast in Deng Xiaoping’s metaphor “crossing the river by groping for stones” – can be seen as an intrinsic part of most Chinese policymaking and reform178. 99 See amongst others Hsiao (2001)151, Gong, Walker and Shi (2007)157, Yip and Hsiao (2008)22, Lei and Lin (2009)153, Wang and colleagues (2009)156, Wagstaff and colleagues (2009)159, Lin, Liu and Chen (2009)165, Sun and colleagues (2009b)167, Yip and Hsiao (2009)168, You and Kobayashi (2009)170, Shi and colleagues (2010)163, Zhang, Yi and Rozelle (2010)171, Babiarz and colleagues (2010)172, Feng (2010)173, and Long and colleagues (2010)175. 100 See amongst others Yip and Hsiao (2008)22, Wagstaff and Lindelow (2008)158, Lei and Lin (2009)153, Wagstaff and colleagues (2009)159, Sun and colleagues (2009a)166, You and Kobayashi (2009)170, and Long and colleagues (2010)175.
62
unanswered question: “(…) has the new scheme attenuated (as it should) or accentuated (as
it shouldn’t) disparities in access and care between the haves and have-nots?”clxxiii.
India, Nepal and Bangladesh
Community Health Insurance in the Indian subcontinent emerged as an effort to improve
access to healthcare and to protect households from catastrophic medical expenditure, most
often embedded in a broader agenda. Of the 49 Indian, Nepalese and Bangladeshi health-
related schemes listed in the 2005 and 2003 inventories of the International Labour
Organizationclxxxi, ,clxxxii clxxxiii, all but three piggybacked onto existing organisations drawn
from a spectrum that ranged from healthcare providers to micro-finance institutions, though
consisted mainly of broad-spectrum development organisations. Typically, the resulting
schemes took the form of NGOs and were able to build on a foundation of trust and financial
capability. Most schemes are of relatively recent origin. Out of the 49 schemes mentioned, 30
were started after 1995 and so coincided with the shift of interest by the micro-finance sector
from micro-credits to micro-insurance. Across the whole Indian subcontinent, CHI focuses
on the poorer sections of society: small farmers, landless labourers, women’s groups, self-
employed vendors, in fact all communities within the informal sector. The most
comprehensive Indian social security report to date – after highlighting that more than 90%
of the Indian workforce is to be found in the informal sector101 – listed more than 50 NGO-
driven social security schemes in India alone, of which 34 provided health insurance102,clxxxiv.
101 In India, the term generally used to denote the informal sector is the ‘unorganised sector’. The NCEUS (National Commission for Enterprises in the Unorganised Sector) in its 2006 social security report used the terms informal sector and ‘unorganised sector’ interchangeably184. 102 In terms of population coverage, the 2006 NCEUS report estimated a total of 5 million people covered by NGO-driven social security schemes of any kind in India (accounting for 1.5% of the workforce in the informal sector). It provided no separate data on population coverage for those schemes offering health insurance. Of all NGO-driven schemes with data available, 15% covered more than 100,000, 40% between 10,000 and 100,000, and 45% less than 10,000 individuals184.
63
Gains in terms of financial coverage in India are modest at country level: the share of
prepayment plans (including CHI, not mandatory health insurance) in private health
expenditure rose from 1.0 to 2.1% between 2000 and 2007, while out-of-pocket expenditure
as a proportion of private health expenditure dropped from 92.2 to 89.9% over the same
periodxlii.
Gains in terms of financial coverage in Nepal are likewise modest at country level: the
share of prepayment plans (including CHI, not mandatory health insurance) in private
health expenditure rose from 0.1 to 0.4% between 2000 and 2007, while out-of-pocket
expenditure as a proportion of private health expenditure dropped from 91.2 to 90.8% over
the same periodxlii.
Financial coverage at country level in Bangladesh leaves much to be desired. Bangladesh in
fact is loosing ground: the share of prepayment plans (including CHI, not mandatory health
insurance) in private health expenditure dropped from 0.1 to 0% between 2000 and 2007;
out-of-pocket expenditure as a proportion of private health expenditure increased from 95.9
to 97.9% over the same periodxlii. Nationwide data on CHI population and service coverage
are not available.
In the Indian subcontinent, scheme enrolment ranges from a few thousand to several
millions. In India, the main benefit on offer is reimbursement of hospital costs, whereas most
of the Nepalese and Bangladeshi schemes focus on first-line healthcare services. One author
has argued that the risk protection in Bangladesh can therefore not be sufficient to reduce
catastrophic health expenditureclxxxv,103. In all the three countries – and particularly in Nepal
and Bangladesh – considerable copayments are often required.
103 The author however recognises that this is a conclusion based on secondary data only185.
64
A closer look at scheme level offers additional insight.
A 2006 overview based on 10 Indian case studies found enrolment rates ranging from 10 to
90%clxxxvi. A study on the ACCORD104 and the SEWA105 schemes reported enrolment rates of
35 and 20% respectively (4,268 and 101,809 members) in 2003106. Significant population
coverage within their target groups was not the only gain. Both schemes reduced the number
of households that would have experienced catastrophic health expenditure (CHE)107
because of hospital expenses, by 57 and 52% respectively. The study also identified ways to
further increase financial coverageclxxxvii.
A second study on the ACCORD scheme showed that members had a hospital admission
rate 2.2 times higher than non-members, independent of pre-existing conditions and other
confounding factorsclxxxviii.
A study on the Yeshasvini scheme108 reported an enrolment rate of 15% (approximately
3,000,000 members) in 2009109. Using rigorous statistical methods to minimise selection bias,
the study analysed the impact of the scheme and documented increased healthcare
104 ACCORD (Action for Community Organisation, Rehabilitation and Development) is a development NGO in Gudalur, Tamil Nadu, South India, active since 1985. The organization gives support to the indigenous population, commonly known as adivasis (literally “original dwellers”) and recognised as scheduled tribes by the Indian constitution. Among several fields of action, health is one, for which ACCORD constituted the ASHWINI (Association for Health Welfare in the Nilgiris), with a hospital (since 1990) and a CHI scheme (since 1992) to make hospital cost accessible. See also http://www.adivasi.net/accord.php and http://www.ashwini.org/105 SEWA (Self-Employed Women’s Association) is a trade union for women in the informal sector, which started in 1972 in Gujarat. Among its many activities, it offers the community of self-employed women an insurance package, comprising life, asset and medical insurance, since 1992. See also http://www.sewa.org/ and http://www.sewainsurance.org/vimosewa.htm 106 Up to 2003 – when it reached an enrolment rate of 20% – SEWA’s CHI scheme targeted its union members only. From 2004 on, CHI was also offered to the members’ husbands. This explains why the 2006 overview reported an enrolment rate of 10% for the SEWA scheme in 2004186: the denominator had suddenly doubled. In 2006, the target population was further expanded to all dependents under age 18. 107 Several definitions and proxies of CHE exist. This study used household expenditure for hospitalization exceeding 10% of the total annual household income as a pragmatic cut-off point187. 108 Yeshasvini – short for Yeshasvini Cooperative Farmers Health Scheme, also known as Yeshasvini Health Insurance, introduced in 2003 by the Karnataka State Department of Co-operation for cooperative rural farmers and informal sector workers, and operated by the Yeshavini Trust – is a nonprofit health insurance scheme with voluntary affiliation. 109 Enrolment rate calculated as proportion of all registered rural cooperative members in Karnataka. Expressed as proportion of the target fixed for the Indian administrative year 2008-2009, this would be 48%. Expressed as proportion of the total rural population in the state, this would be 9%. Yeshasvini started with 1.6 million members in 2003-2004. The number of members augmented every year, except in 2005-2006 when it dropped to 1.5 million.
65
utilisation, better treatment outcomes, reduced out-of-pocket spending and improved annual
income growth among scheme members compared to non-membersclxxxix.
A study on two community-based schemes and one commercial scheme110 in India found a
positive impact on financial protection111 in all three schemes, and more so in the
community-based onescxc.
In Bangladesh, a study showed that the addition of a HmI scheme to Grameen Bank’s micro-
credit scheme was instrumental in increasing the households’ food sufficiency, but not their
income or non-land assets112,cxci.
In terms of organisational practice and institutional design, the experiences in the Indian
subcontinent – and particularly the interplay of civil society innovations and state policy
making in India – have important lessons to offer.
The fact that most CHI schemes are NGO-driven leaves room for organisational diversity.
Where the NGO is also the healthcare provider – as is more frequently the case in Nepal and
Bangladesh than in India113 – the provider usually runs the scheme. Where the NGO has no
healthcare functions, it may act as an insurer for the community and purchase care from
110 The community-based schemes were UpLift Health in urban Pune, Maharashtra, and Nidan in Patna, Bihar. The so-called commercial scheme was that of the Bharatiya Agro Industries Foundation (BAIF) in rural Pune190. Actually the BAIF scheme can be called a partner-agent scheme (see further) from a development NGO186. See also: http://www.upliftindia.org/programmes/community-based-health-mutual-fund/http://www.nidan.in/otherpage.php?page_code_no=15http://www.baif.org.in/aspx_pages/index.asp111 The authors assessed financial protection by measuring ‘financial exposure’, defined as the fraction of cases among the insured with costs above the ceiling, multiplied by the mean actual costs above the ceiling for the insured, divided by the mean income per insured190. 112 The authors saw flaws in the health insurance scheme design and functioning as a possible explanation of the scheme influencing significantly food sufficiency but not other poverty indicators191. 113 Two already mentioned Indian examples are the Students Health Home (SHH) in West Bengal since 1952, and ACCORD’s ASHWINI scheme in Tamil Nadu since 1992. Others are the Voluntary Health Services (VHS) in Tamil Nadu since 1972 (see http://www.vhs-chennai.org/), and the Jowar Rural Health Insurance Scheme (JRHIS) in Maharashtra since 1981186.
66
independent providers114. In a third option, which became increasingly popular in India115,
the NGO purchases insurance – not care – from a formal insurance company. In this so-
called partner-agent116 or linked model, augmented pooling can lead to wider risk
sharingclxxxvi.
This gain may be overshadowed by several drawbacks: where the premium and the benefit
package are based on actuarial calculations, the premium may be prohibitive or the benefit
package too limited. The resulting insurance product cannot be tailored to meet local
conditions, whilst the patient still has to pay up front and reimbursement is often
cumbersome.
However these models are still in evolution: insurance companies are continuously adapting
their products, and several NGOs are either making a case for innovative benefit packages
and/or improving their negotiating capacity with insurance companies and providers.
One salient example of an innovative benefit package within a linked model is that of
Karuna Trust in Karnataka. From 2002 on, this NGO offers poor communities a tailored
health insurance package from the National Insurance Company (NIC, a public insurer).
Only public service providers, which in theory are free of charge, are eligible in the scheme.
Aware of the impact of indirect cost on access to care, Karuna Trust negotiated an insurance
package consisting of compensation for income loss and for drug procurement. It fixed a
premium below the estimated willingness to pay of the below-the-poverty-line (BPL) target
114 Indian examples are the Raigarh Ambikapur Health Association (RAHA) scheme in Chhattisgarh since 1980, and the Dhan (Development of Human Action) Foundation’s Kafamalaj Kalanjiam Vattara Sangam (KKVS) scheme in Tamil Nadu since 2000186. See http://rahaindia.org/ and http://www.dhan.org/115 Two already mentioned Indian examples are the SEWA scheme, originating in Gujarat in 1992, and the Bharatiya Agro Industries Foundation (BAIF) in Pune since 2001. Others are the Navsarjan Trust scheme in Gujarat since 1999, and the Karuna Trust scheme in Karnataka since 2002186 (for the last two schemes, see also http://navsarjan.org/ and http://www.karunatrust.com/wiki/index.php/Health) 116 Classified as a subcategory of HmI by Radermacher and Dror (Institutional options for delivering health microinsurance in Churchill (2006) Protecting the poor: a microinsurance compendium28, pp 401-423; see also page 6 of this document), distinct from their community-based (CHI) subcategory of HmI. Here discussed as a variety of CHI.
67
population, convinced the United Nations Development Programme (UNDP) to subsidise
the premium for the first two years, and the NIC to bear the eventual loss.
While an evaluation of the scheme commissioned by the CGAP117 Working Group on
Microinsurance questioned the Karuna Trust scheme for its dependence on external
subsidiescxcii, the still functioning scheme all the same focused the interest of national policy-
makers on the specific insurance needs of BPL families118.
One salient example of strategic purchasingcxciii,119 within a linked model is that of the SEWA
scheme in Gujarat. From 2004 on, the Self-Employed Women’s Association piloted a
preferred provider system (PPS). Aware of a variety of barriers for their scheme members in
assessing benefits and submitting claimscxciv, SEWA introduced a PPS to facilitate access by
making payment to their members prior to discharge from the hospital, to shift the burden of
compiling a claim to the scheme staff, and to direct members to inpatient facilities of
acceptable quality. Selection of preferred providers was based on the assessment of an
elementary list of structural quality indicators in a set of hospitals already used by SEWA
members, and led to an informal agreement (no contract or memorandum of understanding
was signed) between SEWA Insurance and 16 selected hospitals in 8 rural sub-districts120.
117 Consultative Group to Assist the Poor, a policy and research centre dedicated to advancing financial access (savings, credit, money transfers, insurance) for the world’s poor, World Bank based. See http://www.cgap.org/p/site/c/aboutus/ 118 The National Rural Health Mission’s 2007 framework for developing health insurance programmes declared compensation for loss of wages an imperative for any scheme insuring BPL families206. The Planning Commission’s working group on healthcare financing in its 2006 report mentioned the Karuna Trust scheme as one example of providing protection against catastrophic health expenditure207, p15, and the Planning Commission’s 2007-2012 five-year plan used the example of the Karuna Trust scheme to plead for schemes developed according to local requirements, tailored to the reality of the poor and organised according to their convenience208, p83. 119 The World Health Organization defined strategic purchasing (as opposed to passive purchasing) as “deciding which interventions should be purchased, how, and from whom”193, p97 “to improve the (health) system’s responsiveness and financial fairness”193, p104. 120 SEWA – among structure, process and outcome indicators of service quality – focused on structural indicators because of the difficulty to assess process and outcome indicators in an environment with only rudimentary hospital information systems, and because it felt that it did have not the required technical skills. SEWA decided not to go for formal contracts or memoranda of understanding because this would have been legally impossible with government providers, and deemed not worth the effort with private providers (as the anticipated number of patients per hospital would be small)195.
68
The PPS was successful to the extent that it indeed directed patients to providers of
acceptable quality, and lessened the members’ burden of submitting their claims. Despite the
abolition of retrospective reimbursement however, user fees continued to pose financial
barriers to the insured. An evaluation of the pilot PPS also suggested that SEWA had not the
critical mass of members to express a consolidated purchasing power121,cxcv.
In 2006, SEWA introduced a project with identical objectives and features (prospective
reimbursement plus PPS) – but at that time renamed ‘cashless hospitalisation’122 – in the city
of Ahmedabad. During the first year, members were left the choice between the new
prospective and the traditional retrospective reimbursement. In 2007, ‘cashless
hospitalisation’ became in principle mandatory. Evaluation after 20 months revealed
improved use of low-cost quality care, as in the earlier rural pilot project. Besides, the
average claim cost and the cost of servicing the claims had decreased. SEWA’s ‘cashless
hospitalisation’ thus became a win-win for both the scheme members and the
organisationcxcvi,123.
In most schemes, the providers operate either in the private nonprofit or the private-for-
profit sector, seldom in the public sector. Though most of the operating NGOs lack technical
expertise and a health-systems perspective, many of them have evolved mechanisms to
manage risks, hold providers accountable and increase access to services of acceptable
quality. Interestingly, a recent realist review of eight Indian CHI schemes found that while
all schemes used a mix of measures – that can be classified as long route and exit
121 The authors of the article opined so for the rural environment in which SEWA piloted its PPS. They expected a bigger potential for PPS in urban areas, where SEWA has a larger and more concentrated membership195. 122 Strictly speaking, SEWA replaced retrospective by prospective reimbursement (that is before discharge from the hospital), not by a cashless system. 123 The success of SEWA’s ‘cashless hospitalisation’ was quickly noticed by national policy-makers. When it was still piloted, the Planning Commission’s working group on healthcare financing described it as an “impressive model”207. From 2008 on, an upgraded cashless model would become a foremost feature in the national Rashtriya Swasthya Bima Yojana scheme211.
69
strategies124, ,cxcvii cxcviii – to improve access to quality healthcare, most of these measures were
not very effective. Contrastingly, those schemes125 that strengthened the short route by
empowering their members at the provider-patient interface seem to successfully improve
access to quality healthcarecxcix.
Over the last decade, the lack of technical expertise has been instrumental in establishing
micro-insurance training centres in India (six by 2006), encouraged by donors and the
insurance industry. Whatever gains in technical expertise at the insurers’ side this may have
produced, at the providers’ side quality of care often remains sub-standard and fee-for-
service payment and over-prescribing ubiquitous. Recent research in two Indian CHI
schemes126 found no significant difference in patient satisfaction (as one outcome of service
quality) between scheme members and non-memberscc. In this scenario, the need for
provider regulation is obvious to ensure service quality and to control costs. Still, while India
has come a long way in terms of institutional design to guide organisational practice at the
insurers’ side, it has been much less successful in the domain of provider regulation. India
has not yet found an effective way to regulate its increasingly complex health service
delivery systemcci.
When India opened up its insurance market in 1999, it did so by laying down a cautiously
elaborated institutional framework and establishing its Insurance Regulatory and
124 Building upon the accountability framework proposed by the Word Bank in its 2004 World Development Report197, the authors of the review distinguish short (direct) and long (through politicians and policy-makers) routes of accountability. In line with the earlier conceptual work of Albert Hirschman198, they differentiate further between voice and exit mechanisms. They end up with four underlying mechanisms in the field of CHI: (M1) providing an exit route; (M2) co-producing a long route; (M3) guarding over the long route; and (M4) strengthening the short route by transforming the power imbalance at the provider-patient interface199. 125 This is the case for four of the eight schemes under review: the Action for Community Organisation, Rehabilitation and Development (ACCORD) scheme in Tamil Nadu (see http://www.ashwini.org/) and the Jowar Rural Health Insurance Scheme (JRHIS), UpLift Health (see http://www.upliftindia.org/programmes/community-based-health-mutual-fund/) and the Kagad Kach Patra Kashatakari Panchayat (KKPKP) scheme (see http://www.wastepickerscollective.org/) in Maharashtra199. 126 These schemes were the Action for Community Organisation, Rehabilitation and Development (ACCORD) scheme (see http://www.ashwini.org/) and the Dhan (Development of Human Action) Foundation’s Kafamalaj Kalanjiam Vattara Sangam (KKVS) scheme (see http://www.dhan.org/), both in Tamil Nadu200.
70
Development Authority (IRDA). Main functions of the IRDA – as specified in section 14 of
the 1999 IRDA act – are regulating and promoting any type of insurance, in the best interest
of the policyholdersccii. In the field of CHI (and beyond), the IRDA has made plain use of its
institutional power by imposing on all public and private insurers the duty to serve
incremental quotas of rural and BPL populationscciii.
The Indian government provided a further impetus for pro-poor and community-based
insurance through the establishment of its National Rural Health Mission (NRHM) in 2005.
Up to 2012 the NRHM lays out a vast plan of action aiming at improving the availability of
and access to quality healthcare by the rural poorcciv. While the declared first priority of the
NRHM is to put in place an enabling public health infrastructure, its framework for
implementation also encourages community-based organisations to involve – either as
insurance provider or as third-party administrator – in pro-poor risk pooling
arrangementsccv. The NRHM further elaborated the latter policy in its 2007 framework for
developing health insurance programmesccvi.
Likewise the national Planning Commission set up a working group on healthcare financing.
Among its tasks was to suggest management strategies and process and impact assessment
parameters for community-based health insurance. In its 2006 report, the working group
drew lessons from past experiences, welcomed the call of the NRHM, described the role of
community-based organisations as vital for articulating poor peoples’ needs and stated that
participation of government healthcare providers would be critical for health insurance not
to become a subsidy for private care127,ccvii.
127 Similarly, the NRHM’s framework for developing health insurance programmes stated “that NRHM strategy for health insurance for vulnerable groups is primarily to reduce out of pocket burden of poor families when they go to a government hospital. This will also improve the utilization of government hospitals. The intent of health insurance under NRHM is not to weaken the public system in any manner”206.
71
Subsequently, the Planning Commission’s eleventh five-year plan (2007-2012) encouraged
the development of CHI, tailored to the reality of the poor and organised according to their
convenience. At the same time the Commission recognised that community organisations
still play a “very small (though important) role” in the Indian health system, and that
“coverage of health insurance in India is pathetically limited”ccviii.
The support of Indian government for CHI goes beyond laying down rules for
implementation. In both the case of the Yeshavini and the Karuna Trust scheme, government
subsidies have been essential for survival and successclxxxix,cxcii,128. In the case of the Yeshavini
scheme, the fast enrolment of large numbers of members would not have been possible
without the energetic engagement of a government agencyccix,129.
Indian government welcomes CHI both as a pragmatic way forward130 and as a source of
inspiration. It encourages CHI on the road to universal coverage under the framework of
NRHM (and NUHM, conceivablyccx,131), but also increasingly commits itself to large-scale
financial protection in healthccxi,132, thereby applying lessons learnt from CHI experiences.
The most recent and ambitious example is the Rashtryia Swasthya Bima Yojana (RSBY) scheme
for the BPL population, launched in 2007 and rolled out from April 2008 onccxii,133. Several of
128 These subsidies have different forms and origins: the Yeshavini scheme has received yearly and direct support from the Karnataka state government189; the Karuna Trust scheme has received support from year 3 on through the assumption of losses by the National Insurance Company192. 129 As mentioned beforecv, the Yeshavini scheme belongs to a private nonprofit organisation. But it was the Karnataka State Department of Co-operation (and its Registrar of Co-operative Societies) that provided and enforced the vehicle for communication and enrolment, to the extent that Yeshavini came to be known as a ‘government’ scheme. While the directive involvement of a government agency may have reduced the principle of voluntary affiliation, it was key for fast and broad population coverage208. 130 Particularly the Planning Commission’s eleventh five-year plan makes a definite statement: “These schemes can be implemented where institutional capacity is too weak to organize mandatory nation-wide risk pooling”208, p82. 131 Designed on the lines of the NRHM, a National Urban Health Mission (NUHM) was announced in the 2007-2012 five-year plan193 and a draft mission document was circulated in 2008210. As of end 2010, NUHM has not been launched.
132 Government commitment to health insurance is no new phenomenon in India. The Employees’ State Insurance Scheme (ESIS) for formal-sector workers and their dependents was a first of its kind in 1948; the Central Government Health Scheme (CGHS) for government employees and their families was introduced in 1954. Schemes targeting state- or nation-wide coverage of vulnerable populations however were unknown until the beginning of the present century211. 133 See also http://rsby.gov.in/about_rsby.html
72
its features reflect innovations pioneered by CHI schemes: premiums are subsidised (as in
the Karuna Trust scheme), the system is entirely cashless for the patient at the place and time
of use (perfecting the model introduced in the SEWA scheme), and exclusions are
minimised.
In terms of empowerment, evidence from India is still scarce, yet complementary to the rich
experiences in design and practice. Michielsen and colleagues did both field research onccxiii
and a realist review of the transformative dimension of CHI in Indiacxcix. Focusing on
possible transformative effects of CHI at the patient-provider interface in Mumbai and Pune
slums, they identified how slum dwellers capitalised on their membership and on the
mediation by social workers of CHI schemes to have a say in service qualityccxiii.
Confirmation of this bottom-up empowering effect in their review of eight schemes led the
researchers to criticise the RSBY’s concept of empowerment134 as insufficient, and concluded
that government – in addition to its top-down approach – should tap the potential of
community organisations to transform the RSBY target groups from passive beneficiaries
into active participantscxcix.
Cambodia
Community Health Insurance in Cambodia started with the introduction of the SKY scheme
(Khmer acronym for Health for Our Families) in 1998. SKY was designed and is
implemented by the French NGO GRET (Groupe de recherche et d´échanges technologiques)135 as
134 “Empowering the beneficiary: RSBY provides the participating BPL household with freedom of choice between public and private hospitals and makes him a potential client worth attracting on account of the significant revenues that hospitals stand to earn through the scheme” (http://rsby.gov.in/about_rsby.html). 135 See http://www.gret.org/
73
a complement to a micro-credit scheme run by the same organisation since 1991136. In its
pilot years, SKY suffered from high dropout rates and was redesigned several timesccxiv.
Since 2004 the scheme offers a benefit package including primary and referral care, transport,
and a grant for funeral expenses when neededccxv. By end 2008, SKY was active in four out of
twenty-four rural provinces plus the capital Phnom Penh, and had enrolled nearly 35,000
membersccxvi. By mid 2010, SKY covered 61,000 individuals137. Next to SKY, only a handful of
CHI schemes are operational in Cambodiaccxvii,ccxviii. Most noticeable is the Cambodian
Association for Assistance to Families & Widows (CAAFW) scheme in Banteay Meanchey
province, covering nearly 39,000 individuals by mid 2009ccxix.
Gains in terms of financial and population coverage are of little account at country level:
the share of prepayment plans (including CHI, not mandatory heath insurance) in private
health expenditure remained at 0% between 2000 and 2007, while out-of-pocket expenditure
as a proportion of private health expenditure dropped from 97.1 to 84.7% over the same
periodxlii. According to 2008 data, national population coverage of CHI was still below 1% at
that timeccxx.
The scheme-specific accounts behind the national figures are more encouraging. SKY in 2008
covered between 3 and 14% of its target populations, respectively in districts where the
scheme was most recently introduced and where the scheme was introduced a decade
earlierccxvi. Between 2008 and 2010, SKY almost doubled its number of beneficiariesccxvi,137.
The CAAFW scheme in 2009 covered 31% of its target population in its original location, and
13% in Oddar Meanchey province where it had been introduced half a year earlierccxix.
Between 2005 and 2008, CAAFW had quadrupled its number of CHI beneficiariesccxx. No
136 GRET’s rationale behind the linkage of a micro-credit with a health insurance scheme was the finding that a significant proportion of non-reimbursement of loans in GRET’s micro-credit scheme was due to healthcare expenses214,215. 137 See http://www.gret.org/activites_uk/result_long.asp?pays=37&domaine=&type=&mot=&Submit=Search&cle=1288
74
scheme-specific data on financial coverage are available yet; for SKY the results of an impact
evaluation are expected soonccxxi.
The Cambodian experience with CHI is most interesting in terms of organisational practice
and institutional design.
Each in their own way, both CAAFW and GRET link CHI with a social assistance fund for
free access to healthcare for the extreme poor, known as Health Equity Fund
(HEF)138( , , , , )ccxxii ccxxiii ccxxiv ccxxv ccxxvi .
When CAAFW started its first CHI scheme in 2003, it did so grafting the scheme on a HEF it
had been operating since 2000ccxix. GRET started linking its CHI scheme (SKY, since 1998)
with existing health equity funds in 2005. By 2010, out of SKY’s 61,000 beneficiaries, more
than 18,000 (around 30%) were covered through a HEF137. While health equity funds – just
like community health insurance – are no magic bullet, the articulation of CHI and HEF can
help to overcome the exclusion that the extreme poor despite CHI still face, and appears to
be a plausible step on the road to universal coverageccxxvii.
Both the CAAFW scheme and SKY work exclusively with public healthcare providers, which
not per se offer good quality care in Cambodia. Both developed and successfully negotiated
a range of mechanisms to upgrade quality of care. Particularly the experience of SKY in
adapting its organisational practice is richly documented: in its actual form SKY uses a
capitation system at health centres and first referral hospitals, and a third-payer system at
provincial and national hospitals. It has established contractual relationships with fifty-five
138 Health Equity Funds originated in Cambodia in the year 2000, to overcome the exclusion of the extreme poor from healthcare. The HEF principle of third-party financing for the extreme poor has been adopted – mostly on a small scale – in a number of other countries (see for example the section on Mauritania in this paper). In Cambodia HEF has become part of government policy. For a selection of literature on the Cambodian Health Equity Funds, see Hardeman and colleagues (2004)222, Jacobs and Price (2006)223, Noirhomme and colleagues (2007)224, Annear and colleagues (2008)232, Meessen and colleagues (2008)225, and Bigdeli and Annear (2009)229.
75
health centres, ten first-referral hospitals and five provincial or national hospitals. The terms
of the contracts encompass compliance with intervention protocols, essential drugs stocks,
opening hours and staff presence, and an established referral system. In addition, SKY
monitors patient satisfaction and follows up complaints at village and provider level. As a
consequence, SKY contributed in both raising the quality of care at offer and the utilisation of
public healthcareccxxviii.
These achievements were not easily accomplished nor were they due to organisational
practice alone. For instance, providers used to informal payments were reluctant to give up
this practice. To overcome this and other obstacles, a partnership with and strong
involvement of government at all levels was decisiveccxxviii.
This would not have been possible without the development of an institutional framework
by the central government either. A social health insurance Master Plan was developed in
2003 and adopted in 2005. The plan adheres to a fragmented approach towards universal
coverage139, aiming at coverage by a distinct financing mechanism for different population
groups: compulsory insurance (SHI) for formal-sector workers and their dependents;
voluntary insurance (CHI) for the informal sector workers and their dependents that can
afford it; and social assistance (HEF) for those that cannot afford insurance – first through
external, later through government funding and channelled towards insurance. The plan
includes intermediate and final goals for the government and non-government sectors, and
the establishment of a SHI Committee for intersectoral collaborationccxv,ccxxix.
Following the plan, a new HEF implementation and monitoring framework was established
in 2005 and expanded in 2007, guidelines for implementing CHI saw the light in 2006ccxxix,
139 Comparable to fragmented approaches elsewhere, like in Senegal or Mali in Africa48, or in China.
76
and a National Social Security Fund (NSSF) started functioning in 2008140. Also in 2008, the
Cambodian government released its 2nd Health Strategic Plan (2008-2015). This plan stresses
social health protection – especially for the poor and vulnerable – as a first of five working
principles, describes health financing as a strategic area and announces that by 2015 “the
different elements and institutions of the current health financing system will be combined
under a single strategy”ccxxx. A subsequent Strategic Framework for Health Financing 2008-
2015 outlines a phased approach to bring HEF, CHI and SHI “under a common Social Health
Insurance umbrella” by 2015, to create the path towards universal coverage beyond
2015ccxxxi.
While the 2008-2015 Health Strategic Plan mentions patient empowerment as a way forward
for quality and accountability in service deliveryccxxx, and the 2008-2015 Strategic Framework
for Health Financing explicitly wants to “empower communities to participate in local
policies and decisions that affect their financial access to health services”ccxxxi, actual
Cambodian evidence on dynamics of empowerment and transformation is still scarce. A
2007 study on a selection of CHI schemes and health equity funds found little evidence that
CHI or HEF per se contributed to empower people in their health seeking and treatment
behaviour. The study however perceived an impetus for empowerment through activities of
the implementing NGOs, such as mechanisms for feedback and complaints and the
employment of community liaison officersccxxxii.
Lao Peoples Democratic Republic
140 See http://www.nssf.gov.kh/. Integration of a health component in NSSF’s social protection activities is still pending.
77
The design features of social health protection – including community health insurance and
health equity funds – in Lao People’s Democratic Republic (also known as Laos, hereafter
called Lao PDR) resemble that of Cambodia. Progress is slower, but achievements in terms of
access to services and financial protection are noteworthy.
Gains in terms of financial and population coverage are fair at country level: the share of
prepayment plans (including CHI, not mandatory heath insurance) in private health
expenditure rose from 0 to 0.4% between 2000 and 2007, while out-of-pocket expenditure as
a proportion of private health expenditure dropped from 91.8 to 76.1% over the same
periodxlii.
At country level, population coverage is not impressive. After nine years of piloting, less
than a dozen of schemes now cover 1.7% of all Laotians; or 3.3% of their intended national
target population141. When balanced against their actual local target populations, coverage
reaches 13%. Membership of a CHI scheme significantly increases access to both inpatient
and outpatient services, which is an important step forward considering Lao PDR’s very low
utilisation rates. Despite the fact that CHI members are twice as likely to use health services,
they also enjoy a substantial financial protection. Out-of-pocket payment over 12 months –
including the monthly premiums in the group of the members – is 66% lower for CHI
members than for the non-insuredccxxxiii,142.
Sound organisational practice seems to play a material part in the achievements of CHI in
Lao PDR. The schemes’ comprehensive benefit packages (including primary and referral
care plus drugs obtainable from public structures), effective use of referral care, and a
141 According to Lao policy, CHI targets households in the informal sector not covered by other social protection schemes. This group represents 52% of the total population233. 142 The authors of this study233 contrast this positive outcome with the observed higher OOP payments among the insured in China’s RCMS158.
78
capitation payment system all contribute to the gains in financial protection. At the same
time, revenue generation is considered insufficient and dropout rates are still highccxxxiii.
Progress in institutional design goes hand in hand with organisational practice. Lao PDR
has an active Health Insurance Unit in its Ministry of Health since 2000. A ‘Health Strategy
to the Year 2020’ part of the Sixth National Socio-Economic Development Plan provides the
framework for the development of both community health insurance and health equity
funds, alongside two social health insurance schemesccxxxiv,143. In contrast with the situation
in Cambodia, Lao PDR’s health equity funds are less developed144 than its CHI schemes.
Government plans to link HEF with CHI in the near future, whereby the CHI premium of an
extreme poor would be paid by HEFccxxxiii.
Understandably in this still incipient stage of development, evidence on empowerment and
transformation through CHI is still lacking.
The Philippines
Community health insurance in the Philippines is mainly of historical interest. The first CHI
initiatives in the Philippines were offshoots of community- or cooperative-driven health
programmes in the 1980s, most of them plagued by low enrolment. In addition, several local
government-prompted schemes were set up in the 1990s. At the same time, CHI schemes
(alongside Health Maintenance Organisations) were activated by NGOs with external
assistanceccxxxv. Today, CHI occupies a marginal place in Philippine health financing.
143 Lao PDR’s roadmap is another example of a fragmented approach. The framework projects universal coverage through CHI for the informal workforce, HEFs for the extreme poor, a SHI scheme called Social Security Organisation (SSO) for the private formal sector, and a SHI scheme called Civil Servants’ Scheme (CSS) for the public workforce233,234. By 2009, the four schemes together had reached 11.7% national coverage233. 144 In three aspects: fewer schemes, lower population coverage, and not yet coordinated with other SHP mechanisms233,234.
79
Financial and population coverage at national level leave much to be desired. The
Philippines are actually loosing ground: the share of prepayment plans (including CHI, not
mandatory health insurance) in private health expenditure dropped from 11.1 to 9.8%
between 2000 and 2007, while out-of-pocket expenditure as a proportion of private health
expenditure increased from 77.2 to 83.7% over the same periodxlii. Individual coverage of
CHI schemes is rarely documented.
While CHI schemes in the Philippines typically suffered operational difficulties and
financial problems, the 1995 National Health Insurance Act paved the way for SHI and
created the Philippine Health Insurance Corporation (PhilHealth)145 with the task to ensure
universal coverage by 2010ccxxxvi. In support of the institutional choice for SHI, GTZ
subsequently set up the Social Health Insurance Networking and Empowerment (SHINE)
project to link and frame the disparate CHI efforts within PhilHealth. The project was
discontinued in 2001. PhilHealth today performs well in population coverage (around 80%,
of which 20% poor in a subsidised regime), but less so in depth and height of coverage.
Alongside the national PhilHealth a number of CHI schemes still subsist, most of them at
local or provincial level and/or filling the gaps in outpatient service coverage left by
PhilHealth.
Indonesia
As in the Philippines, CHI in Indonesia is mainly of historical interest. A CHI approach was
adopted as one option among others within the national policy in the late 20th century, but
practically abandoned in the early 21st century.
145 See http://www.philhealth.gov.ph/
80
Gains in terms of financial and population coverage are weak at country level: the share of
prepayment plans (including CHI, not mandatory heath insurance) in private health
expenditure dropped from 6.4 to 4.7% between 2000 and 2007, while out-of-pocket
expenditure as a proportion of private health expenditure dropped from 72.9 to 66.2% over
the same periodxlii.
In 1963, a sickness fund (Dana sakit) arrangement was tried out in one location in Central
Java, within a broader community development effort of a local NGO. The project failed but
was successfully relaunched in 1969, now under the name Dana sehat (Health
Funds)ccxxxvii,ccxxxviii. From 1970 up into the 1990s, the government promoted Dana sehat
community schemes as an alternative form of healthcare financing all over the countryccxxxix.
By the year 2000, Dana sehat had a population coverage of nearly 11%, reaching 23 million
people. This big number should not be misleading. Behind it was a scattered picture of
organisational practice: small schemes (many of them covering less than 500 households)
with insufficient risk pooling, unattractively limited benefit packages and dropout rates as
high as 90%cli,ccxxxix,ccxl. In parallel with the Dana sehat, health maintenance organisations
(Jaminan Pemeliharaan Kesehatan Masyarakat, JPKM) had been launched from 1992 onwards. A
more influential institutional design proved to be the introduction of health benefits cards
for the poor (Kartu sehat) in 1998146. Half a decade later, the Kartu sehat had replaced the Dana
Sehat approach. The Kartu sehat by the time covered just under 20% of the Indonesian
population. Yet only 12% of the cardholders effectively used their card when accessing
health services, and of them 25% reported that they still had to payccxli. In 2004-2005, the
Kartu sehat approach became an integral part of Indonesia’s social security design under the
146 The Kartu sehat – part of a larger social safety net programme – were preceded by another initiative using benefits card (called ‘Letter of Non-affordability’), which applied to health, education and transportation but had a limited health services coverage240 and was less successful239.
81
Asuransi Kesehatan Masyarakat Miskin (Health Insurance for Poor Population) or
Askeskin. A recent impact evaluation of Askeskin found increased utilisation of outpatient
services by the poor, but also increased out-of-pocket spendingccxlii.
Currently, Indonesia projects universal health insurance coverage by 2014ccxlii. It also debates
the pros and cons of different health financing options to reach this goal147. Community
health insurance is no longer on the national agendaccxliii.
147 Basically, three options are under discussion: a NHS-like approach; a unique SHI scheme (in this review called a comprehensive approach); and a combination of government coverage of the poor and mandatory insurance coverage of the others through multiple insurance funds (in this review called a fragmented approach)243.
82
Community health insurance in Latin America
Mutual aid societies have existed in Latin America since the 19th century. Yet Community
Health Insurance is a marginal phenomenon in the continent, especially when compared to
the venerable record of Social Health Insurance in some countries (like Costa Rica148 and
Uruguay149) and of the development of National Health Systems in others (Cuba since
1959150, Brazil since 1988151).
Most Latin American countries however have segmented social protection systems, rarely
covering the whole populationccxliv,ccxlv. Exclusion is common practice, also in healthccxliv,ccxlvi.
It is the preoccupation with the excluded that gave rise to a few CHI initiatives, without
however really taking off. Within a small number of case studies analysed a decade ago by
the International Labour Organization, all led to improved access to healthcare amongst their
target populations, but only a minority were judged to be financially sustainable in the
absence of external fundingccxlvii. Even so, most still exist and new schemes are being
established. In the light of a growing regional commitment to social inclusionccxlviii,ccxlix and to
universal coverage of health servicesccl, it is pertinent to question if and how the scattered
Latin American CHI efforts can contribute to this broader development goals.
148 See – among other sources – http://www.ccss.sa.cr/ and http://www.paho.org/hia/archivosvol2/paisesing/Costa%20Rica%20English.pdf 149 See – among other sources – http://www.bps.gub.uy/ and http://www.paho.org/hia/archivosvol2/paisesing/Uruguay%20English.pdf 150 See – among other sources – http://www.paho.org/hia/archivosvol2/paisesing/Cuba%20English.pdf151 See – among other sources – http://www.paho.org/hia/archivosvol2/paisesing/Brazil%20English.pdf and http://www.who.int/bulletin/volumes/86/4/08-030408.pdf
83
84
Promises and challenges of Community Health Insurance at the crossroads of universal coverage
What is there to be learned from this global overview of CHI?
Our review indicates that the CHI picture today is very patchy, be it in Africa or Asia. In
Latin America, CHI is nowadays hardly relevant and does not deserve much further
discussion.
We observe a great heterogeneity in institutional designs and organisational models for
implementing CHI in both the African and Asian continent. Similarly, we take notice of huge
variation in coverage achieved, in terms of breadth, depth and height.
Except for the cases of Rwanda and Ghana, CHI in sub-Saharan Africa remains a relatively
marginal, although growing phenomenon that currently occupies only a minor role in the
wider endeavour of achieving universal coverage. Coverage at country level indeed rarely
exceeds a few percent. This picture of small coverage at national level deserves nuance: CHI
schemes in Africa are rarely launched on a programmatic nationwide basis. Most schemes
today still are – with the notable exceptions of Rwanda and Ghana in mind – the result of
scattered local project initiatives heavily dependent on support from external organisations.
Marginal national coverage contrasts with the sometimes quite substantial scale achieved at
individual scheme level, illustrating that under certain conditions growth and expansion are
possible. The Bwamanda scheme in the Democratic Republic of Congo is exemplary in that
respect. It shows that things are possible if the context is conducive. One of the important
conditions for CHI to develop and grow is the need for a minimal level of (perceived) quality
of care at the supply side of healthcare. Another one is the requirement of adequate
organisational practice and design within the schemes: responsive to people’s felt needs, but
also financially sound and rational. To take but one example: the a priori quasi systematic
85
introduction of copayments at the time and point of health services utilisation, without any
solid evidence of excessive overconsumption of healthcare by the insured, does not make
much sense. Decades of (mis)application of copayments in low-income settings reveal how
counter-productive this measure has been: it created yet another barrier to the many pre-
existing cultural, social and administrative obstacles that excluded people experience in their
difficult search for quality healthcare.
Ghana and especially Rwanda are powerful examples in Africa showing that political will,
clear action plans, national scope of implementation beyond pilot project settings, existence
of regulatory frameworks, and – last but not least – the unequivocal acceptance of the need
of subsidies to finance partly or totally the premium for the poorest in society are a must.
Under these conditions, CHI in Ghana, or the Mutuelles de Santé in Rwanda, today contribute
significantly to progress towards universal coverage.
The CHI ‘movement’ in Africa has, for far too long, been hostile to a policy of subsidising
insurance premiums for the poorest. The evidence today, however, is straightforward: CHI is
not an (affordable) option for the poorest and the destitute. Unless somebody else pays for
them, they will never be able to join a CHI scheme and remain unprotected. A series of cases
in Cambodia, China and India illustrates the relevance of subsidising schemes, be it with
public or with donor money.
All in all, the Asian picture of CHI is also a variable one. The case of China of course catches
the eye. The New Cooperative Medical Schemes (NCMS) are the expression of a spectacular
revival of CHI in China, with impressive expansion in population coverage in a relatively
short time span. Yet accumulating evidence points to worrisome findings too. There appears
to be only a minor impact of NCMS on financial protection: CHI is paired with steep
increases in out-of-pocket spending, even of higher risk of catastrophic health expenditure.
86
China’s multipronged NCMS experience points to the need for a more rational scheme
design, as well as for stronger provider regulation hand in hand with patient empowerment.
The international evidence today that CHI may be a lever to empower its members in their
relationship with healthcare providers is extremely scanty. Nonetheless, these socio-political
dimensions of CHI have hardly been investigated. Indeed the study of CHI has been – and
still is – largely led by a research community that finds its inspiration in economic and
financial frameworks. The hypothesis that CHI could constitute a lever to weigh on the
responsiveness of the health system and on the quality of healthcare provided remains a
plausible one – given the past experience with sickness funds in early 20th century Europe –
but would need more systematic study in Africa and Asia than is the case today. For CHI or
any other social health protection arrangement to guarantee equitable access, there is need
for empowerment of the most vulnerable groups and individuals. Evidence today – from
women in Nongon and Ahmedabad to poor city dwellers in Nouackchott and Pune –
suggests that CHI can have a positive transformative impact.
There are today no blueprints on how best CHI can be integrated in a national policy
towards universal coverage. The options at hand are path-dependant and subject to the
specificities of the national context. The most frequent picture however seems to be that of a
fragmented approach, with CHI as one of the strategies in a pluralistic environment where
the CHI model coexists with and hopefully complements other financing modalities
targeting specific population groups.
The challenge to streamline and coordinate these various financing options in the broader
perspective of universal coverage remains a huge one. There is need for more thorough
scrutiny and documentation of the set of strategies that have been followed in cases of
successful scaling-up of local CHI initiatives: we know that political will is paramount (it is
87
essential if, eventually, the voluntary character of CHI is to be overcome), but what is the
precise configuration of the institutional and managerial environment that enabled
multiplication and expansion of local initiatives and a solid embedding of CHI in national
policies?
For CHI to maximise its risk pooling potential, schemes can join forces, as has been shown in
Mali and is in progress in several African countries. Or they can make affiliation mandatory,
as did Ghana and Rwanda. Both decisions rely on attitudes towards solidarity. Failing to join
forces – as experienced in Senegal – raises a question mark over the feasibility of voluntary
solidarity. Yet legal enforcement of affiliation is not sufficient to enrol the informal sector, or
to guarantee effective access.
Whatever the role and place of CHI in national health financing systems, it is clear that its
development cannot do without a systemic approach. CHI is about financing healthcare, but
not only about that. It is also about organising and empowering the demand side in the
healthcare system. In addition, development of CHI must go hand in hand with gradually
improving quality at the supply side, with the necessary institutional and regulatory
environment to steer and control provider behaviour. The implementation of CHI therefore
requires a system-wide approach. Similarly, monitoring and evaluation of CHI development
can only benefit from a multi-disciplinary perspective. Evaluation and confirmation of CHI’s
empowering potential would enable decision makers to consider CHI not only as a financing
device, but also as a social investment.
88
REFERENCES
i World Health Assembly (2005) Sustainable health financing, universal coverage and social health insurance. Resolution WHA58.33, Ninth plenary meeting, 25 May 2005 – Committee A, eight report. Geneva: World Health Organization. http://www.who.int/gb/ebwha/pdf_files/WHA58/WHA58_33-en.pdf
ii World Health Assembly (2005) Social health insurance: sustainable health financing, universal coverage and social health insurance. Report by the Secretariat A58/20, 7 April 2005. Geneva: World Health Organization. http://www.who.int/gb/ebwha/pdf_files/WHA58/A58_20-en.pdf
iii World Health Organization (2008) The world health report 2008: primary health care now more than ever. Geneva: World Health Organization. http://www.who.int/whr/2008/whr08_en.pdf
iv Nitayarumphong S (1998) Universal coverage of health care: challenges for the developing countries. In: Nitayarumphong S and Mills A [Editors] Achieving universal coverage of health care. Bangkok, Thailand: Ministry of Public Health; Office of Health Care Reform; pp. 3-24.
v Commission on Social Determinants of Health (2008) Closing the gap in a generation: equity through action on the social determinants of health. Final report of the Commission on Social Determinants of Health (p100). Geneva: World Health Organization. http://whqlibdoc.who.int/publications/2008/9789241563703_eng.pdf
vi Kutzin J (2000) Towards universal health care coverage: a goal-oriented framework for analysis. Health, Nutrition and Population discussion paper. Washington, DC: the World Bank; Human Development Network; Health, Nutrition and Population Family. http://siteresources.worldbank.org/HEALTHNUTRITIONANDPOPULATION/Resources/281627-1095698140167/Kutzin-TowardUniversal-whole.pdf
vii Mills A (2007) Strategies to achieve universal coverage: are there lessons from middle income countries? A literature review commissioned by the Health Systems Knowledge Network. London: London School of Hygiene and Tropical Medicine; Health Economics and Financing Programme. http://web.wits.ac.za/NR/rdonlyres/21C5E48E-702D-4772-979B-779130DC9772/0/Mills2007Universalcoverage.pdf
viii Narbondo P (2006) Reflexiones críticas sobre el universalismo básico. Revista Uruguaya de Ciencia Política 15, 151-172. http://www.fcs.edu.uy/icp/downloads/revista/RUCP15/RevistaICP15-07.pdf
ix Carrin G, Mathauer I, Xu K and Evans D (2008) Universal coverage of health services: tailoring its implementation. Bulletin of the World Health Organization 86(11), 857-863. http://www.who.int/bulletin/volumes/86/11/07-049387.pdf
x Mathauer I and Carrin G (2010) The role of institutional design and organizational practice for health financing performance and universal coverage. Discussion Paper No 5. Geneva: World Health Organization; Department Health Systems Financing; Cluster Health Systems and Services. http://www.who.int/health_financing/documents/dp_e_10_05-instit_uc.pdf
xi Mathauer I and Carrin G (2010) The role of institutional design and organizational practice for health financing performance and universal coverage. Health Policy [in press].
xii North D (1994) Institutional change: a framework of analysis. Economic History, EconWPA. Örebro, Sweden. http://129.3.20.41/eps/eh/papers/9412/9412001.pdf
89
xiii Wallerstein N (2006) Empowerment and related concepts: definitions and dimensions [Annex 1]. In: Wallerstein N [Author] What is the evidence on effectiveness of empowerment to improve health? Copenhagen: WHO Regional Office for Europe; Health Evidence Network; pp 17-19. http://www.euro.who.int/__data/assets/pdf_file/0010/74656/E88086.pdf
xiv De Vos P, De Ceukelaire W, Malaise G, Pérez D, Lefèvre P and Van der Stuyft P (2009) Health through people’s empowerment: a rights-based approach to participation. Health and Human Rights 11(1), 23-35. http://www.hhrjournal.org/index.php/hhr/article/view/126/202
xv Kabeer N, Cook S, Chopra D and Ainsworth (2010) Social protection in Asia: research findings and policy lessons. Programme synthesis report 2010. Sussex, UK; Institute for Development Studies and New Delhi, India; Institute for Human Development: Social Protection in Asia policy-research and network building program. http://www.socialprotectionasia.org/Conf-prgram-pdf/SPA_SynthReport_web.pdf
xvi United Nations Development Programme (2010) Empowerment. In: Klugman J [Lead author] Human Development Report 2010 – The real wealth of nations: pathways to human development. New York: United Nations Development Programme; pp 66-71. http://www.undp.org/publications/hdr2010/en/HDR_2010_EN_Complete.pdf
xvii Sabates-Wheeler R and Devereux S (2007) Social protection for transformation. IDS Bulletin 38(3), 23-28.
xviii Michielsen J, Meulemans H, Soors W, Ndiaye P, Devadasan N, De Herdt T, Verbist G and Criel B (2010) Social protection in health: the need for a transformative dimension [Editorial]. Tropical Medicine and International Health 15(6), 645-658. http://dspace.itg.be/bitstream/10390/6142/1/2010tmih0654.pdf
xix Diop F and Ba A (2010) Mutual health insurance, scaling-up and the expansion of health insurance in Africa. Bethesda, MD: Abt Associates; Health Systems 20/20. http://www.healthsystems2020.org/content/resource/detail/2523/
xx Bennett S, Creese A and Monasch R (1998) Health insurance schemes for people outside formal sector employment. Geneva: World Health Organization; Division of Analysis, Research and Assessment. http://whqlibdoc.who.int/hq/1998/WHO_ARA_CC_98.1.pdf
xxi Criel B, Waelkens M-P, Soors W, Devadasan N and Atim C (2008) Community health insurance in developing countries. In: Heggenhougen K and Quah S [Editors] International Encyclopedia of Public Health. San Diego, CA: Academic Press; pp 782-791.
xxii Yip W and Hsiao (2008) The Chinese health system at a crossroads. Health Affairs 27(2), 460-468. http://content.healthaffairs.org/cgi/reprint/27/2/460
xxiii Witter S and Garshong B (2009) Something old or something new? Social health insurance in Ghana. BMC International Health and Human Rights 9, 20. http://www.biomedcentral.com/content/pdf/1472-698X-9-20.pdf
xxiv Twahirwa A (2008) Sharing the burden of sickness: mutual health insurance in Rwanda. Bulletin of the World Health Organization 86(11), 823-824. http://www.who.int/bulletin/volumes/86/11/08-021108.pdf
90
xxv Waelkens M-P and Criel B (2004) Les mutuelles de santé en Afrique sub-Saharienne: état des lieux et réflexions sur un agenda de recherche. Health, Nutrition and Population discussion paper. Washington, DC: the World Bank; Human Development Network; Health, Nutrition and Population Family. http://siteresources.worldbank.org/HEALTHNUTRITIONANDPOPULATION/Resources/281627-1095698140167/Waelkens-LesMutuelles-whole.pdf
xxvi Dror D and Jacquier C (1999) Micro-insurance: extending health insurance to the excluded. International Social Security Review 52(1), 71-97. http://www3.interscience.wiley.com/cgi-bin/fulltext/119067113/PDFSTART
xxvii International Labour Organization (2000) Health Micro-Insurance: a compendium. Geneva: International Labour Office; Social Security Department; Strategies and Tools against social Exclusion and Poverty programme. http://www.gdrc.org/icm/step.pdf
xxviii Churchill C [Editor] (2006) Protecting the poor: a microinsurance compendium. Geneva: International Labour Office. http://www.munichre-foundation.org/NR/rdonlyres/52FA02DB-B6A4-4DEB-8149-5A64B64D6A68/0/ProtectingthepoorAmicroinsurancecompendiumFullBook.pdf
xxix Institute on Health, Politics and Society in Africa (2005) Access and equity in African health systems. Dakar, Senegal: Council for the Development of Social Science Research in Africa. http://www.codesria.org/IMG/article_PDF/article_a290.pdf
xxx Russell S (1996) Ability to pay for health care: concepts and evidence. Health Policy and Planning 11(3), 219-237. http://heapol.oxfordjournals.org/cgi/reprint/11/3/219
xxxi Gilson L, Kalyalya D, Kuchler F, Lake S, Oranga H and Ouendo M (2000) The equity impacts of community financing activities in three African countries. International Journal of Health Planning and Management 15, 291-317. http://www3.interscience.wiley.com/cgi-bin/fulltext/77002025/PDFSTART
xxxii Ndiaye P (2006) Développement des mutuelles de santé en Afrique: une analyse comparative des approches et de leurs impacts. Research Matters working paper. Ottawa, ON: International Development Research Centre; Governance, Equity and Health program initiative. http://www.idrc.ca/uploads/user-S/11539235471Pascal_Ndiaye_-_Mutuelles_-_juin_2006.pdf
xxxiii Moens F (1990) Design, implementation, and evaluation of a community financing scheme for hospital care in developing countries: a pre-paid health plan in the Bwamanda health zone, Zaire. Social Science & Medicine 30(12), 1319-1327.
xxxiv Oxfam International (2008) Health insurance in low-income countries: where is the evidence that it works? Oxford, UK: Oxfam International. http://publications.oxfam.org.uk/download.asp?dl=http://www.oxfam.org.uk/resources/policy/health/downloads/bp112_health_insurance.pdf
xxxv La Concertation (2008) 10 years serving the African Mutual Health Organization, 1998-2008. Dakar, Senegal: la Concertation. http://www.concertation.org/gimi/concertation/RessFileDownload.do?ressourceId=10211
xxxvi La Concertation (2004) Inventaire des mutuelles de santé en Afrique: Synthèse des travaux de recherche dans 11 pays. Dakar, Senegal: la Concertation. www.ilo.org/gimi/RessFileDownload.do?ressourceId=82
91
xxxvii Ndiaye P, Soors W and Criel B (2007) A view from beneath: Community Health Insurance in Africa [Editorial]. Tropical Medicine and International Health 12(2), 157-161. http://dspace.itg.be/bitstream/10390/1378/1/2007tmih0157.pdfxxxviii Waelkens M-P and Criel B [Editors] (2007) La mise en réseau de mutuelles de santé en Afrique de l’Ouest : l’union fait-elle la force ? Les enseignements d’un colloque international organisé à Nouakchott, Mauritanie, 19 et 20 décembre 2004. Antwerp: ITGPress. http://dspace.itg.be/bitstream/10390/1620/1/9789076070308.pdf
xxxix Union Economique et Monétaire Ouest Africaine (2009) Règlement Nº07/2009/CM/UEMOA portant réglementation de la mutualité sociale au sein de l’UEMOA. Dakar, Senegal: Union Economique et Monétaire Ouest Africaine; le Conseil des Ministres. http://www.uemoa.int/actualite/2009/CM26062009/reglement_07_2009_CM_UEMOA.pdf
xl Union Economique et Monétaire Ouest Africaine (2009) Communiqué de presse de la reunion ordinaire du conseil des ministres de l’union, Dakar, le 26 juin 2009. Dakar, Senegal: Union Economique et Monétaire Ouest Africaine; le Conseil des Ministres. http://www.uemoa.int/actualite/2009/CM26062009/com_presse_26062009.pdf
xli Projet d’Appui aux Systèmes de Santé des Régions Médicales de Kaolack et Fatick (2008) Une couverture assurance-maladie pour les élèves. Kaolack, Senegal: ASSRMKF. http://assrmkf.sn/spip.php?article40
xlii World Health Organization (2010) World Health Statistics 2010. Geneva: World Health Organization; Department of Health Statistics and Informatics. http://www.who.int/whosis/whostat/EN_WHS10_Full.pdf
xliii Belgische Technische Coöperatie/Coopération technique belge (2008) Programme d’Appui au développement de la MicroAssurance Santé au Sénégal (PAMAS 2008-2011) : dossier technique et financier. Brussels: BTC/CTB. http://www.btcctb.org/files/web/tenders/Dossier%20technique%20et%20financier%20Appui%20au%20développement%20de%20la%20micro%20assurance%20santé%20au%20Sénégal.pdf
xliv Jütting J (2003) Do community-based health insurance schemes improve poor people’s access to health care? Evidence from rural Senegal. World Development 32(2), 273-288. http://www.oecd.org/dataoecd/48/46/24670935.pdf
xlv Chankova S, Sulzbach S and Diop F (2008) Impact of mutual health organizations: evidence from West Africa. Health Policy and Planning 23(4), 264-276. http://heapol.oxfordjournals.org/cgi/reprint/23/4/264
xlvi République du Sénégal (2004) Plan stratégique de développement des mutuelles de santé au Sénégal. Dakar, Senegal: Ministère de la Santé. http://www.sante.gouv.sn/IMG/pdf/planstrategique.pdf
xlvii Wade A (2007) La mise en réseau des mutuelles de santé au Sénégal. In: Waelkens M-P and Criel B [Editors] La mise en réseau de mutuelles de santé en Afrique de l’Ouest : l’union fait-elle la force ? Les enseignements d’un colloque international organisé à Nouakchott, Mauritanie, 19 et 20 décembre 2004. Antwerp: ITGPress; pp 65-84. http://dspace.itg.be/bitstream/10390/1620/1/9789076070308.pdf
xlviii Letourmy A (2010) How to implement health insurance in sub-Saharan Africa? Background paper for the European Report on Development (ERD 2010) conference Promoting resilience through social
92
protection in sub-Saharan Africa, 28-30 June 2010. Dakar, Senegal: ERD 2010 regional conference. http://erd.eui.eu/media/2010/Letourmy%20ERD.pdf
xlix Faye O and Paul E (2009) The opportunities of the global crisis for social policy enhancement in Senegal. Geneva: United Nations Research Institute for Social Development. http://www.unrisd.org/80256B3C005BD6AB/httpNetITFramePDF?ReadForm&parentunid=116BFCFB01D47A66C12576A200567079&parentdoctype=eventauxiliarypage&netitpath=80256B3C005BD6AB/(httpAuxPages)/116BFCFB01D47A66C12576A200567079/$file/FayePaul.pdf
l Witter S, Armar-Klemesu M and Dieng T (2008) National fee exemption schemes for deliveries: comparing the recent experiences of Ghana and Senegal. In: Richard F, Witter S and De Brouwere V [Editors] Reducing financial barriers to obstetric care in low-income countries. Studies in Health Services Organisation & Policy, 24. Antwerp: ITGPress; pp 167-198. http://dspace.itg.be/bitstream/10390/2576/2/2008shso0167.pdf
li Witter S, Dieng T, Mbengue D, Moreira I and De Brouwere V (2010) The national free delivery and caesarean policy in Senegal: evaluating processes and outcomes. Health Policy and Planning 25(5), 384-392. http://heapol.oxfordjournals.org/cgi/reprint/25/5/384
lii Pereznieto P (2009) Social protection to tackle child policy in Senegal. ODI Project Briefing No 26. London: Overseas Development Institute. http://www.odi.org.uk/resources/download/3793-odi-project-briefing-english.pdf
liii Smith K and Sulzbach S (2008) Community-based insurance and access to maternal health services: evidence from three West African countries. Social Science & Medicine 66(12), 2460-2473.
liv Ouimet M-J, Fournier P, Diop I and Haddad S (2007) Solidarity or financial sustainability: an analysis of the values of community-based health insurance subscribers and promoters in Senegal. Canadian Journal of Public Health 98(4), 341-346. http://journal.cpha.ca/index.php/cjph/article/viewFile/885/885
lv Togo L, Ponzio N, Bouaré M and Ouattara O (2004) Rapport Mali sur la couverture sociale de la maladie. Background document for the colloquium L’amélioration de l’accès aux services de santé en Afrique francophone : le role de l’assurance, 28-29 April 2004. Paris: Collège des Economistes de la Santé; Réseau Economie de la Santé dans les Pays en Développement. http://www.ces-asso.org/docs/WBI_IMA_Rapport_Mali.pdf
lvi Letourmy A and Ouattara O (2006) L’assurance maladie obligatoire au Mali : discussion d’un processus en cours. In: Dussault G, Fournier P and Letourmy A [Editors] L’assurance maladie en Afrique francophone : améliorer l’accès aux soins et lutter contre la pauvreté. Health, Nutrition and Population discussion paper. Washington, DC: the World Bank, Human Development Network; Health, Nutrition and Population Family; pp 229-262. http://www-wds.worldbank.org/external/default/WDSContentServer/WDSP/IB/2006/08/30/000160016_20060830113122/Rendered/PDF/37149.pdf
lvii Fischer K, Sissouma I and Hathie I (2006) L’Union Technique de la Mutualité Malienne, Mali. CCGAP Working Group on Microinsurance: Good and Bad Practices, Case Study No 23. Geneva: International Labour Office; Social Finance Programme. http://www.ilo.org/wcmsp5/groups/public/---ed_emp/documents/publication/wcms_122479.pdf
93
lviii Ministère du Développement Social, de la Solidarité et des Personnes Agées & Bureau Régional de l’UNICEF pour l’Afrique de l’Ouest et du Centre (2009) La protection sociale et les enfants en Afrique de l’Ouest et du Centre : le cas du Mali. Bamako, Mali: Ministère du Développement Social, de la Solidarité et des Personnes Agées; Direction Nationale de la Protection Sociale et de l’Economie Solidaire. http://www.unicef.org/wcaro/wcaro_mali_Protection_sociale_Mali_FR_s.pdf
lix Miller Franco L, Pathé Diop F, Burgert C, Gamble Kelly A, Makinen M and Tidiane Simpara C (2008) Effects of mutual health organizations on use of priority health services in urban and rural Mali: a case-control study. Bulletin of the World Health Organization 86(11), 830-838. http://www.who.int/bulletin/volumes/86/11/08-051045.pdf
lx Miller Franco L, Pathé Diop F, Burgert C, Gamble Kelly A, Makinen M and Tidiane Simpara C (2008) Effects of mutual health organizations on use of priority health services in urban and rural Mali: a case-control study: Appendices listed in the Bulletin of the World Health Organization 86(11), November 2008. http://www.urc-chs.com/news/WHO_Appendices.pdf
lxi Soors W (2006) Evaluation de la Mutuelle de Santé de Cinzana. Background document for Akory A, Soors W and Douabélé C (2006) Evaluation du Projet Intégré de Santé et d’Activités Mutualistes (PISAM) dans la commune de Cinzana (cercle de Ségou). External evaluation report. Bamako, Mali: Novartis Foundation for Sustainable Development [Unpublished; available from the author upon request [email protected]].
lxii Criel B, Atim C, Basaza R, Blaise P and Waelkens M-P (2004) Community health insurance (CHI) in sub-Saharan Africa: researching the context [Editorial]. Tropical Medicine and International Health 9(10), 1041-1043. http://dspace.itg.be/bitstream/10390/1212/1/2004tmih1041.pdf
lxiii Togola A and Gerber D (2007) An evaluation of Malian civil society’s role in governance. New York: Open Society Institute; Africa Governance Monitoring and Advocacy Project (AfriMAP). http://www.afrimap.org/english/images/paper/Mali_Civil_Society(fin).pdf
lxiv Audibert M and de Roodenbeke E (2005) Utilisation des services de santé de premier niveau au Mali : analyse de la situation et perspectives. Washington, DC: the World Bank; Région Afrique; Département du développement humain. http://www-wds.worldbank.org/external/default/WDSContentServer/WDSP/IB/2005/09/28/000160016_20050928112649/Rendered/PDF/33643a10MLI0sante0AFR0Human0dev0no087.pdf
lxv Koné Y, Ouattara O and Balique H (1998) Un projet de création d’une mutuelle en zone rurale au Mali : la mutuelle des cotonniers de Nongon. Background paper for the conference Les mutuelles de santé en Afrique : concept importé ou réalité émergente ? Expérience et perspectives, 27-29 May 1998. Paris: Centre International de l’Enfance et de la Famille.
lxvi Coulibaly S, Desplat D, Koné Y, Nimaga K, Dugas S, Farnarier G, Sy M, Balique H, Doumbo O and Van Dormael M (2007) Une médecine rurale de proximité : l’expérience des médecins de campagne au Mali. Education for Health 20(2), 47. http://www.educationforhealth.net/publishedarticles/article_print_47.pdf
lxvii Letourmy A and Pavy-Letourmy A (2005) La micro-assurance de santé dans les pays à faible revenu. Notes et documents, No 26. Paris: Agence Française de Développement; Département de la Recherche. http://www.afd.fr/jahia/webdav/site/afd/users/administrateur/public/publications/notesetdocuments/ND-26.pdf
lxviii Ouattara O and Sissouma I (2007) La mise en réseau des mutuelles de santé : l’expérience de l’Union Technique de la Mutualité malienne. In: Waelkens M-P and Criel B [Editors] La mise en réseau de mutuelles
94
de santé en Afrique de l’Ouest : l’union fait-elle la force ? Les enseignements d’un colloque international organisé à Nouakchott, Mauritanie, 19 et 20 décembre 2004. Antwerp: ITGPress ; pp 49-64. http://dspace.itg.be/bitstream/10390/1620/1/9789076070308.pdf
lxix Union Technique de la Mutualité Malienne (2009) Le role de la Mutualité Malienne dans l’extension de la couverture maladie au plus grand nombre : synthèse generale des travaux. Séminaire organisé par l’UTM en partenariat avec la Mutualité Française, Bamako, 24-26 février 2009. Bamako, Mali: Union Technique de la Mutualité Malienne.
lxx Ministère de la Santé du Mali (2009) La mutualité malienne s’engage dans l’amélioration des services de santé. http://www.sante.gov.ml/index.php?option=com_content&view=article&id=776:la-mutualite-malienne-sengage-dans-lamelioration-des-services-de-sante&catid=78&Itemid=120
lxxi McIntyre D, Garshong B, Mtei G, Meheus F, Thiede M, Akazili J, Ally M, Aikins M, Mulligan J-A and Goudge J (2008) Beyond fragmentation and towards universal coverage: insights from Ghana, South Africa and the United Republic of Tanzania. Bulletin of the World Health Organization 86(11), 871-876. http://www.who.int/bulletin/volumes/86/11/08-053413.pdf
lxxii Soors W and Criel B (2009) L’interface entre mutuelles, qualité de l’offre et demande : résultats de deux etudes de cas au Mali rural. Presentation at the Forum 2009 de la Concertation. Yaoundé, Cameroon: Palais des Congrès. http://www.concertation.org/gimi/concertation/ShowNews.do?nid=5843[embedded as Plénière 2 Thème 2]
lxxiii Samaké S, Keita B and Mariko S (2007) Santé de la mère. In: Samaké S, Traoré S, Ba S, Dembélé E, Diop M, Mariko S and Libité P [Editors] Enquête Démographique et de Santé du Mali 2006. Bamako, Mali: Ministère de la Santé; Direction Nationale de la Statistique de l’Informatique; Cellule de Planification et de Statistique – Calverton, MD: Macro International. http://www.measuredhs.com/pubs/pdf/FR199/FR199.pdf
lxxiv Sylla M, Barry A, Diallo Y, Yobouno M, Bongono M, Keita A and Condé S (2002) Test du modèle de Mutuelle Communautaire d’Aire de Santé (MUCAS) en Guinée Forestière 1996-2000 (Rapport avril 2000). In: Criel B, Barry A and von Roenne F [Editors] Le projet PRIMA en Guinée Conakry : une expérience d’organisation de mutuelles de santé en Afrique rurale. Brussels: Medicus Mundi Belgium; pp 17-120.
lxxv Ndiaye P, Kaba S, Kourouma M, Barry A, Barry A and Criel B (2008) MURIGA in Guinea: an experience of community health insurance focused on obstetric risks. In: Richard F, Witter S and De Brouwere V [Editors] Reducing financial barriers to obstetric care in low-income countries. Studies in Health Services Organisation & Policy, 24. Antwerp: ITGPress; pp 117-148. http://dspace.itg.be/bitstream/10390/2480/1/2008redu0117.pdf
lxxvi Barry A (2007) La plateforme des opérateurs d’appui aux mutuelles de santé en Guinée. In: Waelkens M-P and Criel B [Editors] La mise en réseau de mutuelles de santé en Afrique de l’Ouest : l’union fait-elle la force ? Les enseignements d’un colloque international organize à Nouakchott, Mauritanie, 19 et 20 décembre 2004. Antwerp: ITGPress; pp 99-105. http://dspace.itg.be/bitstream/10390/1620/1/9789076070308.pdf
lxxvii Gbamou C (2007) La mise en réseau des mutuelles de santé en Guinée : l’expérience du CIDR et de l’Union des mutuelles de la Guinée forestière. In: Waelkens M-P and Criel B [Editors] La mise en réseau de mutuelles de santé en Afrique de l’Ouest : l’union fait-elle la force ? Les enseignements d’un colloque international organize à Nouakchott, Mauritanie, 19 et 20 décembre 2004. Antwerp: ITGPress; pp 85-97. http://dspace.itg.be/bitstream/10390/1620/1/9789076070308.pdf
95
lxxviii Criel B and Waelkens M-P (2003) Declining subscriptions to the Maliando Mutual Health Organisation in Guinea-Conakry (West Africa): what is going wrong? Social Science & Medicine 57(7), 1205-1219.
lxxix Wiegandt A, Van der Vennet J, Diallo A and Criel B (2002) La perception des professionels de santé de la MUCAS Maliando : la difficulté de l’approche partenariale. In: Criel B, Barry A and von Roenne F [Editors] Le projet PRIMA en Guinée Conakry : une expérience d’organisation de mutuelles de santé en Afrique rurale. Brussels: Medicus Mundi Belgium; pp 207-238.
lxxx Criel B, Diallo A, Van der Vennet J, Waelkens M-P and Wiegandt A (2005) La difficulté du partenariat entre professionels de santé et mutualistes : le cas de la mutuelle de santé Maliando en Guinée-Conakry. Tropical Medicine and International Health 10(5), 450-463. http://dspace.itg.be/bitstream/10390/1213/1/2005tmih0450.pdf
lxxxi Criel B, Blaise P and Ferette D (2006) Mutuelles de santé en Afrique et qualité de soins dans les services : une interaction dynamique. In: Dussault G, Fournier P and Letourmy A [Editors] L’assurance maladie en Afrique francophone : améliorer l’accès aux soins et lutter contre la pauvreté. Health, Nutrition and Population discussion paper. Washington, DC: the World Bank, Human Development Network; Health, Nutrition and Population Family; pp 229-262. http://www-wds.worldbank.org/external/default/WDSContentServer/WDSP/IB/2006/08/30/000160016_20060830113122/Rendered/PDF/37149.pdf
lxxxii Djan B, Tafsir C and von Roenne F (2002) L’économie du centre de santé de Yendé [Annexe 8]. In: Criel B, Barry A and von Roenne F [Editors] Le projet PRIMA en Guinée Conakry : une expérience d’organisation de mutuelles de santé en Afrique rurale. Brussels: Medicus Mundi Belgium; pp 112-118.
lxxxiii Dong H, De Allegri M, Gnawali D, Souares A and Sauerborn R (2009) Drop-out analysis of community-based health insurance membership at Nouna, Burkina Faso. Health Policy 92(2), 174-179.
lxxxiv Gnawali D, Pokhtrel S, Sié A, Sanon M, De Allegri M, Souares A, Dong H and Sauerborn R (2009) The effect of community-based health insurance membership on the utilization of modern health care services: evidence from Burkina Faso. Health Policy 90(2), 214-222.
lxxxv Wereldsolidariteit. Landenfiche Burkina Faso. http://www.wereldsolidariteit.be/wereldwijd/53-landenfiches/87-burkina-faso
lxxxvi Ridde V, Haddad S, Yacoubou M and Yacoubou I (2010) Exploratory study of the impacts of mutual health organizations on social dynamics in Benin. Social Science & Medicine 71(3), 467-474.
lxxxvii Gerkens M and Jadin K (2009) Bezoek aan zeven projecten van het Belgisch Overlevingsfonds in Benin (15 tot 22 maart 2009): verslag namens de parlementaire delegatie / Visite de sept projets du Fonds belge de survie au Bénin (15 au 22 mars 2009) : rapport fait au nom de la délégation parlementaire [Doc 52 2074/001]. Brussels: Belgische Kamer van volksvertegenwoordigers/Chambre des représentants de Belgique. http://www.jadin.be/medias/pdf/propositions-de-loi/260609%20-%20Benin.pdf
lxxxviii Sery J-P and Letourmy A (2006) Couverture du risqué maladie en Afrique francophone : état des lieux, defies et perspectives. In: Dussault G, Fournier P and Letourmy A [Editors] L’assurance maladie en Afrique francophone : améliorer l’accès aux soins et lutter contre la pauvreté. Health, Nutrition and Population discussion paper. Washington, DC: the World Bank, Human Development Network; Health, Nutrition and Population Family; pp 197-225. http://www-
96
wds.worldbank.org/external/default/WDSContentServer/WDSP/IB/2006/08/30/000160016_20060830113122/Rendered/PDF/37149.pdf
lxxxix Soors W, Waelkens M-P and Criel B (2008) Community health insurance in sub-Saharan Africa: opportunities for improving access to emergency obstetric care? In: Richard F, Witter S and De Brouwere V [Editors] Reducing financial barriers to obstetric care in low-income countries. Studies in Health Services Organisation & Policy, 24. Antwerp: ITGPress; pp 149-164. http://dspace.itg.be/bitstream/10390/2482/1/2008redu0149.pdf
xc Plate-forme des Promoteurs de Mutuelles de Santé au Cameroun/Platform of Promoters of Mutual Health Organisations in Cameroon (2006) Plan stratégique pour la promotion et le développement des mutuelles de santé au Cameroun/Strategic plan for the promotion and development of mutual health organizations in Cameroon. Yaoundé, Cameroon: Promuscam. http://www.plateformecm.org/plateforme/index2.php?cat=planstrategique http://www.plateformecm.org/plateforme/en/index2.php?cat=strategicplan
xci Ahawo A and Kopnang P (2007) How mutual health organizations (MHOs) and microinsurance schemes contribute to the fight against HIV/AIDS in Cameroon. G-NEWS [Social Security Extension and Microinsurance Newsletter] 2, 5. http://www.ilo.org/gimi/RessFileDownload.do?ressourceId=3112
xcii Bâ A (2007) Les mutuelles de santé et l’accès à l’hôpital : perspective de la mutuelle de santé de Dar-Naïm. In: Waelkens M-P and Criel B [Editors] La mise en réseau de mutuelles de santé en Afrique de l’Ouest : l’union fait-elle la force ? Les enseignements d’un colloque international organize à Nouakchott, Mauritanie, 19 et 20 décembre 2004. Antwerp: ITGPress; pp 99-105. http://dspace.itg.be/bitstream/10390/1620/1/9789076070308.pdf
xciii Criel B, Bâ A, Kane F, Noirhomme M and Waelkens M-P (2010) Une experience de protection sociale pour les plus démunis : le fonds d’indigence de Dar-Naïm en Mauritanie. Studies in Health Services Organisation & Policy, 26. Antwerp: ITGPress. http://dspace.itg.be/bitstream/10390/2895/2/2010maur0026.pdf
xciv Criel B, Van Damme W, Meessen B and Ir P (2008) A comparative analysis of public social assistance systems in Belgium and health equity funds in Cambodia: an overview of lessons learned. In: In: Meessen B, Pei X, Criel B and Bloom G [Editors] Health and social protection: experiences from Cambodia, China and Lao PDR. Studies in Health Services Organisation & Policy, 23. Antwerp: ITGPress; pp 347-381. http://www.povill.com/download/A%20comparative%20analysis%20of%20public%20social%20assistance%20systems.pdf
xcv Atim C, Grey S, Apoya P, Anie S and Aikins M (2001) A survey of health financing schemes in Ghana. Bethesda, MD: Abt Associates; Partners for Health Reformplus. http://www.healthsystems2020.org/files/1517_file_Tech013_fin.pdf
xcvi Sulzbach S, Garshong B and Owusu-Banahene G (2005) Evaluating the effects of the national health insurance act in Ghana: baseline report. Bethesda, MD: Abt Associates; Partners for Health Reformplus. http://www.abtassociates.com/reports/national_health_insurance_ghana_1205.pdf
xcvii Gyapong J, Garshong B, Akazili J, Aikins M, Agyepong I and Nyonator F (2007) Critical analysis of Ghana’s health system with a focus on equity challenges and the national health insurance. SHIELD
97
workpackage 1 report. Accra, Ghana: Ghana Health Service; Health Research Unit; Strategies for Health Insurance for Equity in Less Developed countries project. http://web.uct.ac.za/depts/heu//SHIELD/reports/Ghana1.pdf
xcviii Agyepong I and Adjei S (2008) Public social policy development and implementation: a case study of the Ghana national health insurance scheme. Health Policy and Planning 23(2), 150-160. http://heapol.oxfordjournals.org/content/23/2/150.full.pdf
xcix Jehu-Appiah C, Aryeetey G, Spaan E, Agyepong I and Baltussen (2010) Efficiency, equity and feasibility of stategies to identify the poor: an application to premium exemptions und national health insurance in Ghana. Health Policy 95(2), 166-173.
c Alfers L (2009) The Ghana national health insurance scheme. WIEGO social protection case study. Manchester, UK: Women in Informal Employment Globalizing and Organizing. http://www.wiego.org/program_areas/social_protect/ghana_nhis_case_study_draft.pdf
ci Marriott A (2009) Your money or your life: will leaders act now to save lives and make health care free in poor countries? Oxford, UK: Oxfam International. http://www.oxfam.org/sites/www.oxfam.org/files/bp-your-money-or-your-life.pdf
cii Agyepong I, Nabyonga Orem J and Hercot D (2010) When the ‘non-workable ideological best’ becomes the enemy of the ‘imperfect but workable good’ [Viewpoint]. Tropical Medicine and International Health [in press].
ciii Sarpong N, Loag W, Fobil J, Meyer C, Adu-Sarkodie Y, May J and Schwarz N (2010) National health insurance coverage and socio-economic status in a rural district in Ghana. Tropical Medicine and International Health 15(2), 191-197. http://onlinelibrary.wiley.com/doi/10.1111/j.1365-3156.2009.02439.x/pdf
civ Basaza R, Pariyo G and Criel B (2009) What are the emerging features of community health insurance schemes in East Africa? Risk Management and Healthcare Policy 2, 47-53. http://www.dovepress.com/getfile.php?fileID=4955
cv Kamuzora P and Gilson L (2007) Factors influencing implementation of the community health fund in Tanzania. Health Policy and Planning 22(2), 95-102. http://heapol.oxfordjournals.org/content/22/2/95.full.pdf
cvi Msuya J, Jütting J and Asfaw A (2004) Impact of community health insurance schemes on health care provision in rural Tanzania. ZEF discussion papers on development policy no. 82. Bonn: Universität Bonn; Zentrum für Entwicklungsforschung – Center for Development Research. http://www.zef.de/fileadmin/webfiles/downloads/zef_dp/zef_dp82.pdf
cvii Msuya J, Jütting J and Asfaw A (2007) Impact of community health funds on the access to health care: empirical evidence from rural Tanzania. International Journal of Public Administration 30(8), 813-833.
cviii Mtei G and Mulligan J-A (2007) Community health funds in Tanzania: a literature review. Ifakara, Tanzania: Ifakara Health Research and Development Centre; Consortium for Research on Equitable Health Systems programme. http://www.crehs.lshtm.ac.uk/downloads/publications/Community%20health%20funds%20in%20Tanzania.pdf
98
cix Tanzanian German Programme to Support Health (2007) Impact of new ways of public private partnerships in expanding social health insurance to workers in the informal sector. Dar es Salaam, Tanzania: Tanzanian German Programme to Support Health; Health financing component. http://www.tgpsh.or.tz/uploads/media/Impact_of_PPP_in_expanding_social_health_insurance_to_workers_of_the_infomral_sector.doc.pdf
cx Mtei G, Mulligan J, Ally M, Palmer N and Mills A (2007) An assessment of the health financing system in Tanzania. Report on SHIELD work package 1. Ifakara, Tanzania: Ifakara Health Research and Development Centre; Strategies for Health Insurance for Equity in Less Developed countries project. http://web.uct.ac.za/depts/heu//SHIELD/reports/Tanzania1.pdf
cxi Carrin G, Doetinchem O, Kirigia J, Mathauer I and Musango L (2008) Social health insurance: how feasible is its expansion in the African region? DevISSues 10(2), 7-9. http://www.iss.nl/content/download/14393/137519/file/DevISSues10_2_nov_08.pdf
cxii Kenya Community Based Health Financing Association (2008) The role of KCBHFA in the transformation of National Hospital Insurance Fund (NHIF) to National Social Health Insurance Fund (NSHIF). Nairobi, Kenya: Kenya Community Based Health Financing Association. http://www.kcbhfa.org/training%20materials/role%20of%20kcbhfa%20in%20transformation%20of%20nhif.pdf
cxiii Basaza R, Criel B and Van der Stuyft P (2007) Low enrolment in Uganda community health insurance schemes: underlying causes and policy implications. BMC Health Services Research 7, 105. http://dspace.itg.be/bitstream/10390/1161/1/2007bhsr0105.pdf
cxiv Basaza R, Criel B and Van der Stuyft P (2008) Community health insurance in Uganda: why does enrolment remain low? A view from beneath. Health Policy 87(2), 172-184.
cxv Waelkens M-P (2010) Design health insurance in Uganda: evaluation of a Cordaid funded CHI programme in Kabale and Masaka diocese. Nkozi, Uganda: Uganda Martyrs University.
cxvi Basaza R, Criel B and Van der Stuyft P (2010) Community health insurance amidst abolition of user fees in Uganda: the view from policy makers and health service managers. BMC Health Services Research 10, 33. http://dspace.itg.be/bitstream/10390/2978/1/2010bhsr0033.pdf
cxvii Nabyonga Orem J and Mukeiki Zikusooka C (2010) Health financing reform in Uganda: how equitable is the proposed national health insurance scheme? International Journal for Equity in Health 9, 23. http://www.equityhealthj.com/content/pdf/1475-9276-9-23.pdf
cxviii Musango L, Doetinchem O and Carrin G (2009) De la mutualisation du risque maladie à l’assurance universelle : expérience du Rwanda. Discussion paper 1/2009. Geneva: World Health Organization; Department Health Systems Financing; Cluster Health Systems and Services. http://www.who.int/health_financing/documents/dp_f_09_01-mutualisation_rwa.pdf
cxix Roenen C and Criel B (1997) La mutuelle du Kanage : leçons à tirer d’un échec. Une expérience d’un système d’assurance maladie pour soins de santé à l’hôpital de Murunda au Rwanda. L’Enfant en Milieu Tropical 228, 29-39.
99
cxx Schneider P, Diop F and Leighton C (2001) Pilot testing prepayment for health services in Rwanda: results and recommendations for policy directions and implementation. Technical report no. 66; executive summary. Cambridge, MA: Abt Associates Inc. http://www.abtassociates.com/reports/ES-2001230338148_27336.pdf
cxxi Schneider P and Diop F (2001) Synopsis of results on the impact of community-based health insurance on financial accessibility to health care in Rwanda. Health, Nutrition and Population discussion paper. Washington, DC: the World Bank; Human Development Network; Health, Nutrition and Population Family. http://siteresources.worldbank.org/HEALTHNUTRITIONANDPOPULATION/Resources/281627-1095698140167/Schneider-SynopsisofResults-whole.pdf
cxxii Inyarubuga H (2007) Les mutuelles de santé au Rwanda : une force pragmatique de mutualisation de risque lié à la maladie. Kigali, Rwanda: République du Rwanda; Ministère de la Santé; Cellule technique d’appui aux mutuelles de santé. http://www.cbhirwanda.org.rw/index.php?option=com_docman&task=doc_download&gid=6&Itemid=38
cxxiii Ministère de la Santé (2004) Politique de développement des mutuelles de santé au Rwanda. Kigali, Rwanda: République du Rwanda; Ministère de la Santé. http://www.grandslacs.net/doc/4061.pdf
cxxiv Basinga P, Sekabaraga C and Soucat A (2008) Scaling up innovations in services delivery of primary health care: case of Rwanda. In: Three decades of primary health care: reviewing the past and defining the future [Panel session: primary health care: past achievement, future challenges and responses: five country case studies]. Bangkok, Thailand: Prince Mahidol Award Conference 2008; pp 91-108. http://www.pmaconference.mahidol.ac.th/index.php?option=com_docman&task=doc_download&gid=14&Itemid
cxxv National Institute of Statistics of Rwanda (2009) Statistical Yearbook: 2009 Edition. Kigali, Rwanda: Republic of Rwanda; Ministry of Finance and Economic Planning; National Institute of Statistics. http://statistics.gov.rw/images/PDF/Statistical%20Yearbook.pdf
cxxvi National Institute of Statistics of Rwanda (2008) Rwanda service provision assessment survey 2007. Kigali, Rwanda: Republic of Rwanda; Ministry of Finance and Economic Planning; National Institute of Statistics – Calverton, MD: Macro International. http://statistics.gov.rw/images/PDF/RSPA%202007%20FINAL%20REPORT-ENGLISH.pdf
cxxvii Logie D, Rowson M and Ndagije (2008) Innovations in Rwanda’s health system: looking to the future. The Lancet 372(9634), 256-261.
cxxviii National Institute of Statistics of Rwanda (2007) EICV poverty analysis for Rwanda’s economic development and poverty reduction strategy [Final report]. Kigali, Rwanda: Republic of Rwanda: Ministry of Finance and Economic Planning; National Institute of Statistics. http://statistics.gov.rw/images/PDF/2-EICV%20Living%20Standards%20Analysis_English.pdf
cxxix Saksena P, Fernandes Antunes A, Xu K, Musango L and Carrin G (2010) Impact of mutual health insurance on access to health care and financial risk protection in Rwanda. Discussion paper 3/2010. Geneva: World Health Organization; Department Health Systems Financing; Cluster Health Systems and Services. http://www.who.int/health_financing/documents/dp_e_10_03-mut_frp_rwa.pdf
100
cxxx Saksena P, Fernandes Antunes A, Xu K, Musango L and Carrin G (2010) Mutual health insurance in Rwanda: evidence on access to care and financial risk protection. Health Policy [in press].
cxxxi Kalk A, Groos N, Karasi J-C and Girrbach E (2010) Health systems strengthening through insurance subsidies: the GFATM experience in Rwanda [Viewpoint]. Tropical Medicine and International Health 15(1), 94-97. http://onlinelibrary.wiley.com/doi/10.1111/j.1365-3156.2009.02424.x/pdf
cxxxii Schmidt J-O, Mayindo J and Kalk A (2006) Thresholds for health insurance in Rwanda: who should pay how much? Tropical Medicine and International Health 11(8), 1327-1333. http://onlinelibrary.wiley.com/doi/10.1111/j.1365-3156.2006.01661.x/pdf
cxxxiii Kalk A (2008) Rwanda’s health system: some concerns [Correspondence]. The Lancet 372(9651), 1729.
cxxxiv Baza A, Hakizimana A, Hanson K, Kwizera F and van der Geest (1993) Health insurance and the Bamako initiative in Burundi: value for money? The International Journal of Health Planning and Management 8(2), 129-135.
cxxxv Arhin D (1994) The health card insurance scheme in Burundi: a social asset or a non-viable venture. Social Science & Medicine 39(6), 861-870.
cxxxvi Arhin-Tenkorang D (2004) Experience of community health financing in the African region. In: Preker A and Carrin G [Editors] Health financing for poor people: resource mobilization and risk sharing. Washington, DC: the World Bank; pp 157-197.
cxxxvii Kippenberg J, Sahokwasama J and Amon J (2007) Detention of insolvent patients in Burundian hospitals. Health Policy and Planning 23(1), 14-23. http://heapol.oxfordjournals.org/content/23/1/14.full.pdf
cxxxviii Cliffe S and Ndikumana D (1999) Burundi poverty note: prospects for social protection in a crisis economy. Report nº 1709-BU. Washington, DC: the World Bank; Africa Region; Country Department 9. http://www.grandslacs.net/doc/2301.pdf
cxxxix Ministère de la Santé Publique (2006) Plan National de Développement Sanitaire 2006-2010. Bujumbura, Burundi: République du Burundi; Ministère de la Santé Publique. http://www.minisante.bi/index.php?option=com_rokdownloads&view=file&Itemid=86&id=53
cxl Kamwenubusa T, Nicobaharaye O, Niyonkuru D and Munyandekwe O (2010) Etude comparative des systèmes de protection sociale au Rwanda et au Burundi. Brussels: Wereldsolidariteit – Solidarité Mondiale.
cxli Atim C and Criel B (2004) Faisabilité de la mise en oeuvre de mutuelles de santé en République Démocratique du Congo. Mission report. Kinshasa, Democratic Republic of Congo: Ministry of Public Health and Belgian Technical Cooperation – Antwerp: Institute of Tropical Medicine.
cxlii Criel B and Kegels G (1997) A health insurance scheme for hospital care in Bwamanda district, Zaire: lessons and questions after 10 years of functioning. Tropical Medicine and International Health 2(7), 654-672. http://onlinelibrary.wiley.com/doi/10.1046/j.1365-3156.1997.d01-349.x/pdf
cxliii C.D.I.-Bwamanda Integrale Streekontwikkeling (2010) Nieuwsbrief juni 2010. Heverlee, Belgium: C.D.I.-Bwamanda-België.
101
http://www.cdibwamanda.com/website/Documents/newsLetters/nieuwsbrief16/NBjuni2010RESULTATEN2009.pdf
cxliv Criel B, Van der Stuyft and Van Lerberghe (1999) The Bwamanda hospital insurance scheme: effective for whom? A study of its impact on hospital utilization patterns. Social Science & Medicine 48(7), 897-911.
cxlv Chenge F and Bottieau E (2010) Evaluation du projet DGCD OS2 Santé 2008-2010 pour le CDI-Bwamanda. Heverlee, Belgium: CDI Bwamanda België. http://www.cdibwamanda.com/website/paginas/Publicaties/Evaluaties/2010Gezondheidszorg.pdf
cxlvi Programme National de Promotion des Mutuelles de Santé (2009) Plan 2009-2011 : les grandes lignes. Synthèse des travaux de l’atelier du 13 au 16/02/2009. Kinshasa, Democratic Republic of Congo: Ministry of Public Health, PNPMS. http://www.coopami.org/fr/countries/countries_partners/drc/projects/2009/pdf/20090202report.pdf
cxlvii Devadasan N (2006) The Student’s Health Home – Kolkata. In: Devadasan N [Editor] Community Health Insurance in India: a compilation of case studies. Ahmedabad, India: Friends of Women’s World Banking; pp 34-37. http://www.iphindia.org/wp/wp-content/uploads/2010/10/CHI-India-compilation-Deva-2006-FWWB.pdf
cxlviii Hsiao W (1999) The Chinese health care system: lessons for other nations. Social Science & Medicine 41(8), 1047-1055.
cxlix Liu Y, Rao K and Hu S (2002) People’s Republic of China: toward establishing a rural health protection system. Manila, the Philippines: Asian Development Bank. http://www.adb.org/Documents/Reports/PRC_Rural_Health_Protection_System/prc_rural_health.pdf
cl Bogg L, Hengjin D, Keli W, Wenwei C and Diwan V (1996) The cost of coverage: rural health insurance in China. Health Policy and Planning 11(3), 238-252. http://heapol.oxfordjournals.org/cgi/reprint/11/3/238
cli Hsiao W (2001) Unmet health needs of two billion: is community financing a solution? Health, Nutrition and Population discussion paper. Washington, DC: the World Bank; Human Development Network; Health, Nutrition and Population Family. http://whoindia.org/LinkFiles/Commision_on_Macroeconomic_and_Health_03_07.pdf
clii Carrin G, Ron A, Hui Y, Hong W, Tuohong Z, Licheng Z, Shuo Z, Yide Y, Jiaying C, Qicheng J, Zhaoyang Z, Jun Y and Xuesheng (1999) The reform of the rural cooperative medical system in the People’s Republic of China: interim experience in 14 pilot counties. Social Science & Medicine 48(7), 961-972. http://www.who.int/health_financing/documents/reform_rural_china_1999.pdf
cliii Lei X and Lin W (2009) The new cooperative medical scheme in rural China: does more coverage mean more service and better health? Health Economics 18(S2), S25-46. http://onlinelibrary.wiley.com/doi/10.1002/hec.1501/pdf
102
cliv Yip W and Hanson K (2009) Purchasing health care in China: experiences, opportunities and challenges. In: Chernichovsky D and Hanson K [Editors] Innovations in health system finance in developing and transitional economies. Bingley, UK: Emerald; pp 197-220.
clv Yi H, Zhang L, Singer K, Rozelle S and Atlas S (2009) Health insurance and catastrophic illness: a report on the new cooperative medical system in rural China. Health Economics 18(S2), S119-S127. http://onlinelibrary.wiley.com/doi/10.1002/hec.1510/pdf
clvi Wang H, Yip W, Zhang L and Hsiao W (2009) The impact of rural mutual health care on health status: evaluation of a social experiment in rural China. Health Economics 18(S2), S65-S82. http://onlinelibrary.wiley.com/doi/10.1002/hec.1465/pdf
clvii Gong S, Walker A and Shi G (2007) From Chinese model to U.S. symptoms: the paradox of China’s health system. International Journal of Health Services 37(4), 651-672.
clviii Wagstaff A and Lindelow M (2008) Can insurance increase financial risk? The curious case of health insurance in China. Journal of Health Economics 27(4), 990-1005. http://repository.upenn.edu/cgi/viewcontent.cgi?article=1019&context=gansu_papers
clix Wagstaff A, Yip W, Lindelow M and Hsiao W (2009) China’s health system and its reform: a review of recent studies. Health Economics 18(S2), S7-S23. http://onlinelibrary.wiley.com/doi/10.1002/hec.1518/pdf
clx Wagstaff A, Lindelow M, Wang S and Zhang S (2009) Reforming China’s rural health system. Washington, DC: the World Bank; Human Development Network; Directions in Development series. http://depositfiles.com/en/files/kr2uz9rm6
clxi Xu Y, Zhang X and Zhu X (2008) Medical financial assistance in rural China: policy design and implementation. In: Meessen B, Pei X, Criel B and Bloom G [Editors] Health and social protection: experiences from Cambodia, China and Lao PDR. Studies in Health Services Organisation & Policy, 23. Antwerp: ITGPress; pp 295-317. http://www.povill.com/download/Medical%20financial%20assistance%20in%20rural%20China%20-%20policy%20design%20and%20implementation.pdf
clxii Zhang J, Ma X, Decoster K, Tang X, Gao X and Meessen B (2008) Helping the poor against major illness: a comparative analysis of medical financial assistance in four countries of China. In: Meessen B, Pei X, Criel B and Bloom G [Editors] Health and social protection: experiences from Cambodia, China and Lao PDR. Studies in Health Services Organisation & Policy, 23. Antwerp: ITGPress; pp 319-345. http://dspace.itg.be/bitstream/10390/2391/2/2008heal0319.pdf
clxiii Shi W, Chongsuvivatwong V, Geater A, Zhang J, Zhang H and Brombal D (2010) The influence of the rural health security schemes on health utilization and household impoverishment in rural China: data from a household survey of western and central China. International Journal for Equity in Health 9, 7. http://www.equityhealthj.com/content/pdf/1475-9276-9-7.pdf
clxiv Hao Y, Wu Q, Zhang Z, Gao L, Ning N, Jiao M and Zakus D (2010) The impact of different benefit packages of medical financial assistance scheme on health service utilization of poor population in rural China. BMC Health Services Research 10, 170. http://www.biomedcentral.com/content/pdf/1472-6963-10-170.pdf
103
clxv Lin W, Liu G and Chen G (2009) The urban resident basic medical insurance: a landmark reform towards universal coverage in China. Health Economics 18(S2), S83-S96. http://onlinelibrary.wiley.com/doi/10.1002/hec.1500/pdf
clxvi Sun X, Jackson S, Carmichael G and Sleigh A (2009a) Prescribing behaviour of village doctors under China’s new cooperative medical scheme. Social Science & Medicine 68(10), 1775-1779.
clxvii Sun X, Jackson S, Carmichael G and Sleigh A (2009b) Catastrophic medical payment and financial protection in rural China: evidence from the new cooperative medical scheme in Shandong province. Health Economics 18(1), S103-S119. http://onlinelibrary.wiley.com/doi/10.1002/hec.1346/pdf
clxviii Yip W and Hsiao W (2009) Non-evidence-based policy: how effective is China’s new cooperative medical scheme in reducing medical impoverishment? Social Science & Medicine 68(2), 201-209.
clxix Wagstaff A, Lindelow M, Jun G, Ling X and Juncheng Q (2009) Extending health insurance to the rural population: an impact evaluation of China’s new cooperative medical scheme. Journal of Health Economics 28(1), 1-19.
clxx You X and Kobayashi (2009) The new cooperative medical scheme in China. Health Policy 91(1), 1-9. http://www.sciencedirect.com/science/article/B6V8X-4V936HH-1/2/b6c788a74c686d2e706cf54e4c5e98c7
clxxi Zhang L, Yi H and Rozelle S (2010) Good and bad news from China’s new cooperative medical scheme. IDS Bulletin 41(4), 95-106.
clxxii Babiarz KS, Miller G, Yi H, Zhang L and Rozelle S (2010) New evidence on the impact of China’s New Rural Cooperative Medical Scheme and its implications for rural primary healthcare: multivariate difference-in-difference analysis. BMJ 341, c5617. http://www.bmj.com/content/341/bmj.c5617.full.pdf
clxxiii Feng Z (2010) Chinese healthcare in rural areas [Editorial]. BMJ 341, c5254. http://www.bmj.com/content/341/7779/Editorials.full.pdf
clxxiv Bogg L, Wang K and Diwan V (2002) Chinese maternal health in adjustment: claim for life. Reproductive Health Matters 10(20), 95-107.
clxxv Long Q, Zhang T, Hemminki E, Tang X, Huang K Xiao S and Tolhurst R (2010) Utilisation, contents and costs of prenatal care under a rural health insurance (New Co-operative Medical System) in rural China: lessons from implementation. BMC Health Services Research 10, 301. http://www.biomedcentral.com/content/pdf/1472-6963-10-301.pdf
clxxvi Bogg L, Huang K, Long Q, Shen Y and Hemminki E (2010) Dramatic increase of Cesarean deliveries in the midst of health reforms in rural China. Social Science & Medicine 70(10), 1544-1549. http://groups.stakes.fi/NR/rdonlyres/727B7CB2-2CA0-422E-8099-BB2B105756A9/0/Boggetal_3_SSM_2010.pdf
clxxvii Leung J (2006) The emergence of social assistance in China. International Journal of Social Welfare 15(3), 188-198. http://onlinelibrary.wiley.com/doi/10.1111/j.1468-2397.2006.00434.x/pdf
clxxviii Hope N, Yang D and Li D [Editors] (2003) How far across the river? Chinese policy reform across the millennium. Stanford, CA: Stanford University Press.
104
clxxix Pan X, Zhang Y, Xu L, Huang J and Zhao Q (2009) An analysis of farmers’ perception of the new cooperative medical system in Liaoning Province, China. BMC Health Services Research 9, 230. http://www.biomedcentral.com/content/pdf/1472-6963-9-230.pdf
clxxx Bloom G, Kanjilai B and Peters D (2008) Regulating health care markets in China and India. Health Affairs 27(2), 952-963. http://content.healthaffairs.org/cgi/reprint/27/4/952
clxxxi International Labour Organization (2005) India: an inventory of micro-insurance schemes. Geneva: International Labour Office; Social Security Department; Strategies and Tools against social Exclusion and Poverty programme. http://www.ilo.org/gimi/amin/RessShowRessource.do?ressourceId=84
clxxxii International Labour Organisation (2003) An inventory of micro-insurance schemes in Nepal. Kathmandu, Nepal: International Labour Office. http://www.ilo.org/public/english/protection/socsec/step/download/714p1.pdf
clxxxiii International Labour Organization (2003) Micro-insurers: inventory of micro-insurance schemes in Bangladesh. Geneva: International Labour Office; Social Security Department; Strategies and Tools against social Exclusion and Poverty programme. http://www.ilo.org/gimi/amin/RessShowRessource.do?ressourceId=83
clxxxiv National Commission for Enterprises in the Unorganised Sector (2006) Social security for unorganized workers: report. New Delhi, India: National Commission for Enterprises in the Unorganised Sector. http://nceuis.nic.in/Social%20Security%20-%20Cover%20and%20index.htm
clxxxv Werner W (2009) Micro-insurance in Bangladesh: risk protection for the poor? Journal of Health, Population and Nutrition 27(4), 563-573. http://www.banglajol.info/index.php/JHPN/article/view/3402/2849
clxxxvi Devadasan N, Ranson K, Van Damme W, Acharya A and Criel B (2006) The landscape of community health insurance in India: an overview based on 10 case studies. Health Policy 78(2), 224-234. http://www.iphindia.org/wp/wp-content/uploads/2010/08/The-landscape-of-CHI-India.pdf
clxxxvii Devadasan N, Criel B, Van Damme W, Ranson K and Van der Stuyft P (2007) Indian community health insurance schemes provide partial protection against catastrophic health expenditure. BMC Health Services Research 7, 43. http://dspace.itg.be/bitstream/10390/1244/1/2007bhsr0043.pdf
clxxxviii Devadasan N, Criel B, Van Damme W, Manoharan S, Sankara Sarma P and Van der Stuyft P (2010) Community health insurance in Gudalur, India, increases access to hospital care. Health Policy and Planning 25(2), 145-154. http://heapol.oxfordjournals.org/content/25/2/145.full.pdf
clxxxix Aggarwal A (2010) Impact evaluation of India’s ‘Yeshasvini’ community-based health insurance programme. Health Economics 19(S1), 5-35. http://onlinelibrary.wiley.com/doi/10.1002/hec.1605/pdf
cxc Dror D, Radermacher R, Khadikar S, Schout P, Hay F-X, Singh A and Koren R (2009) Microinsurance: innovations in low-cost health insurance. Health Affairs 28(6), 1788-1798. http://content.healthaffairs.org/cgi/reprint/28/6/1788
cxci Hamid S, Roberts J and Mosley P (2010) Can micro health insurance reduce poverty? Evidence from Bangladesh. Sheffield Economic Research Paper Series 2010001. Sheffield: University of Sheffield;
105
Department of Economics. http://www.microfinancegateway.org/gm/document-1.9.43777/Can%20Micro%20Health%20Insurance%20Reduce%20Poverty.pdf
cxcii Radermacher R, van Putten-Radermacher O, Müller V, Wig N and Dror D (2005) Karuna Trust, Karnataka, India. CGAP Working Group on Microinsurance: Good and Bad Practices, case study No. 19. Geneva: International Labour Office; Social Finance Programme. http://www.ilo.org/wcmsp5/groups/public/---ed_emp/documents/publication/wcms_122475.pdf
cxciii World Health Organization (2000) The World Health Report 2000: health systems: improving performance. Geneva: World Health Organization. http://www.who.int/whr/2000/en/whr00_en.pdf
cxciv Sinha T, Ranson K, Chatterjee M, Acharya A and Mills A (2006) Barriers to accessing benefits in a community-based insurance scheme: lessons learnt from SEWA insurance, Gujarat. Health Policy and Planning 21(2), 132-142. http://heapol.oxfordjournals.org/content/21/2/132.full.pdf
cxcv Ranson K, Sinha T, Ganghi F, Jayswal R and Mills A (2006) Helping members of a community-based health insurance scheme access quality inpatient care through development of a preferred provider system in rural Gujarat. The National Medical Journal of India 19(5), 274-282. http://www.nmji.in/archives/Volume_19_5_Sep_Oct_2006/Medicine_and_Society/Me&So_5.pdf
cxcvi Kharva R (2008) Vimo SEWA’s cashless hospitalization: a case study for micro health insurance. IRDA Journal 6(2), 31-34. http://www.irdaonline.org/irdacontent/journals/irda_jan08.pdf
cxcvii The World Bank (2003) World Development Report 2004: making services work for poor people. Washington, DC: the World Bank and Oxford University Press. http://www-wds.worldbank.org/external/default/WDSContentServer/IW3P/IB/2003/10/07/000090341_20031007150121/Rendered/PDF/268950PAPER0WDR02004.pdf
cxcviii Hirschman A (1970) Exit, voice, and loyalty: responses to decline in firms, organizations and states. Cambridge, MA: Harvard University Press.
cxcix Michielsen J, Criel B, Devadasan N, Soors W, Wouters E and Meulemans H (2010) Contextual factors for improving access to quality care in India: the issue of health insurance. International Journal for Quality in Health Care [submitted].
cc Devadasan N, Criel B, Van Damme W, Lefèvre P, Manoharan S and Van der Stuyft P (2010) The effect of community health insurance schemes on patient satisfaction: evidence from India. Indian Journal of Medical Research [in press].
cci Peters D and Muraleedharan (2008) Regulating India’s health services: To what end? What future? Social Science & Medicine 66(2), 2131-2144.
ccii Government of India (1999) Insurance Regulatory and Development Authority Act. New Delhi, India: Government of India; Parliament of India. http://www.irda.gov.in/ADMINCMS/cms/frmGeneral_Layout.aspx?page=PageNo108&flag=1&mid=Insurance%20Laws%20etc.%3E%3EActs
cciii Government of India (2007) Duties, powers and functions of IRDA [IRDA/GEN/03/2007]. Hyderabad, India: Insurance Regulatory and Development Authority. http://www.irda.gov.in/ADMINCMS/cms/NormalData_Layout.aspx?page=PageNo101&mid=1.2
106
cciv National Rural Health Mission (2005) National Rural Health Mission (2005-2012) mission document. New Delhi, India: Government of India, Ministry of Health & Family Welfare. http://mohfw.nic.in/NRHM/Documents/Mission_Document.pdf
ccv National Rural Health Mission (2005) Framework for implementation 2005-2012. New Delhi, India: Government of India, Ministry of Health & Family Welfare. http://jknrhm.com/Guideline/Frame_Work.pdf
ccvi National Rural Health Mission (2007) Framework for developing health insurance programmes: some suggestions for states. New Delhi, India: Government of India, Ministry of Health & Family Welfare. http://mohfw.nic.in/NRHM/Documents/framework_for_health_insurance.pdf
ccvii Ministry of Health & Family Welfare (2006) Report of the working group on health care financing including health insurance for the 11th five year plan. New Delhi, India: Government of India, Ministry of Health & Family Welfare. http://planningcommission.nic.in/aboutus/committee/wrkgrp11/wg11_rphfw3.pdf
ccviii Planning Commission, Government of India (2008) Eleventh five year plan, 2007-12; volume II: social sector. New Delhi, India: Oxford University Press. http://planningcommission.nic.in/plans/planrel/fiveyr/11th/11_v2/11th_vol2.pdf
ccix Kuruvilla S and Liu M (2007) Health security for the rural poor? A case study of a health insurance scheme for rural farmers and peasants in India. International Social Security Review 60(4), 3-21. http://digitalcommons.ilr.cornell.edu/cgi/viewcontent.cgi?article=1167&context=articles
ccx Urban Health Division (2008) National Urban Health Mission (2008-2012): meeting the health challenges of urban population especially the urban poor (with special focus on urban slums). Draft for circulation. New Delhi, India: Government of India, Ministry of Health & Family Welfare. http://www.uhrc.in/downloads/Reports/NUHM-Draft.pdf
ccxi Nagpal S, Devadasan N and Jain N (2008) Government initiatives in health insurance: health security for all. IRDA Journal 6(2), 25-30. http://www.irdaonline.org/irdacontent/journals/irda_jan08.pdf
ccxii Devadasan N and Swarup A (2008) Rashtriya Swasthya Bima Yojana: an overview. IRDA Journal 6(4), 33-36. http://www.irdaonline.org/irdacontent/journals/irda_mar08.pdf
ccxiii Michielsen J, John D and Chaudhuri L (2009) Researching the transformative dimensions of social protection in health (SHP): transformations in the power relation between providers and female slum dwellers in Mumbai and Pune (India). Antwerp: Universiteit Antwerpen and VLIR/UOS. http://webh01.ua.ac.be/cello/index.php?pg=7&idrec=380&act=2&filter=8
ccxiv Poursat C (2004) Quelles articulations entre politique de santé et micro-assurance ? Réflexions à partir du projet de micro-assurance santé du Gret au Cambodge. Les documents de travail de la Direction scientifique, Coopérer aujourd’hui nº 37. Paris: Groupe de recherche et d’échanges technologiques. http://www.sky-cambodia.org/downloads/Case_Study_Cambodia_French.pdf and http://www.gret.org/ressource/pdf/07007.pdf
ccxv International Labour Organization (2008) Cambodia: SKY health insurance scheme. Series Social Security Extension Initiatives in East Asia, paper 20. Bangkok, Thailand: International Labour
107
Organization Subregional Office for East Asia. http://www.ilo.org/public/english/region/asro/bangkok/events/sis/download/paper29.pdf
ccxvi International Labour Organization (2009) Cambodia: GRET “SKY” health insurance programme. Micro-insurance brief. New Delhi, India: International Labour Organization Subregional Office for South Asia. http://www.ilo.org/gimi/RessShowRessource.do?ressourceId=6194
ccxvii Levine D and Gardner R (2008) Briefing paper: health care in Cambodia. Berkeley, CA: University of California, Haas School of Business. http://www.afd.fr/jahia/Jahia/site/afd/lang/en/pid/28601
ccxviii Annear P, Bigdeli M, Eang R and Jacobs B (2008) Providing access to health services for the poor: health equity in Cambodia. In: Meessen B, Pei X, Criel B and Bloom G [Editors] Health and social protection: experiences from Cambodia, China and Lao PDR. Studies in Health Services Organisation & Policy, 23. Antwerp: ITGPress; pp 189-226. http://www.povill.com/download/Providing%20access%20to%20health%20services%20for%20the%20poor%20-%20health%20equity%20in%20Cambodia.pdf
ccxix International Labour Organization (2009) Cambodia: Cambodian Association for Assistance to Families & Widows (CAAFW) community health insurance. Micro-insurance brief. New Delhi, India: International Labour Organization Subregional Office for South Asia. http://www.ilo.org/gimi/RessShowRessource.do?ressourceId=6190
ccxx Minnik V (2009) Community based health insurance (CBHI) in Cambodia. Presentation at the Delegation of the European Commission workshop “Repercussions of the economic downturn in rural Cambodia”, Siem Reap, Cambodia, 2-3 April 2009. http://www.delkhm.ec.europa.eu/en/events/april-2-3-2009/09-Community-Based-Health-Insurance.pdf
ccxxi Levine D, Hema N and Ramage I (2007) Insuring health: testing the effectiveness of micro-health insurance to promote economic wellbeing for the poor. BASIS brief 2007-05. Madison, WI: University of Wisconsin-Madison; Department of Agricultural and Applied Economics; BASIS Research Program on Poverty, Inequality and Development. http://www.basis.wisc.edu/publications_ama/pubs_project.html
ccxxii Hardeman W, Van Damme W, van Pelt M, Ir P, Kimvan H and Meessen B (2004) Access to health care for all? User fees plus a Health Equity Fund in Sotnikum, Cambodia. Health Policy and Planning 19(1), 22-32. http://heapol.oxfordjournals.org/content/19/1/22.full.pdf
ccxxiii Jacobs B and Price N (2006) Improving access for the poorest to public sector health services: insights from Kirivong operational health district in Cambodia. Health Policy and Planning 21(1), 27-39. http://heapol.oxfordjournals.org/content/21/1/27.full.pdf
ccxxiv Noirhomme M, Meessen B, Griffiths F, Ir P, Jacobs B, Thor R, Criel B and Van Damme W (2007) Improving access to hospital care for the poor: comparative analysis of four health equity funds in Cambodia. Health Policy and Planning 22(4), 246-262. http://heapol.oxfordjournals.org/content/22/4/246.full.pdf
ccxxv Meessen B, Chheng K, Decoster K, Heng T and Chap S (2008) Can public hospitals be pro-poor? The health equity fund experience in Cambodia. In: Meessen B, Pei X, Criel B and Bloom G [Editors] Health and social protection: experiences from Cambodia, China and Lao PDR. Studies in Health Services Organisation & Policy, 23. Antwerp: ITGPress; pp 469-490.
108
http://www.povill.com/download/Can%20public%20hospitals%20be%20pro-poor%20-%20The%20health%20equity%20fund%20experience%20in%20Cambodia.pdf
ccxxvi Bigdeli M and Annear P (2009) Barriers to access and the purchasing function of health equity funds: lessons from Cambodia. Bulletin of the World Health Organization 87(7), 560-564. http://www.who.int/bulletin/volumes/87/7/08-053058.pdf
ccxxvii Jacobs B, Bigdeli M, van Pelt M, Ir P, Salze C and Criel B (2008) Bridging community-based health insurance and social protection for health care – a step in the right direction of universal coverage? [Editorial]. Tropical Medicine and International Health 13(2), 140-143. http://dspace.itg.be/bitstream/123456789/2221/2/2008tmih0140.pdf
ccxxviii Duffau A and Díaz Pedregal V (2009) To what extent does non profit private micro health insurance help improve public health care? Field Actions Science Reports 3 http://factsreports.revues.org/index360.html?file=1
ccxxix Annear P (2009) Cambodia: developing a strategy for social health protection. In: Promoting sustainable strategies to improve access to health care in the Asian and Pacific region. Bangkok, Thailand: United Nations Economic and Social Commission for Asia and the Pacific; pp 1-35. http://www.unescap.org/esid/hds/pubs/2529/pub2529.pdf
ccxxx Kingdom of Cambodia (2008) Health strategic plan 2008-2015: accountability, efficiency, quality, equity. Phnom Penh, Cambodia: Ministry of Health. http://www.moh.gov.kh/files/dphi/HSP0815En.pdf
ccxxxi Kingdom of Cambodia (2008) Strategic framework for health financing 2008-2015. Phnom Penh, Cambodia: Ministry of Health; Department of Planning and Health Information; Bureau of Health Economics and Financing. http://www.who.int/health_financing/documents/cam_frmwrk.pdf http://www.who.int/health_financing/documents/cam_frmwrk-annex_1.pdf http://www.who.int/health_financing/documents/cam_frmwrk-annex_2.pdf
ccxxxii Annear P, Bigdeli M, Eang R and James P (2007) Study of financial access to health services for the poor in Cambodia – Phase 2: In-depth analysis of selected case studies. Phnom Penh, Cambodia: Ministry of Health. http://www.who.int/health_financing/documents/cam-healthaccess-study-phase2.pdf
ccxxxiii Alkenbrack S, Lindelow M and Jacobs B (2010) Community-based health insurance in Lao P.D.R.: understanding enrolment and impacts [Policy note]. Vientiane, Lao People’s Democratic Republic: Lao People’s Democratic Republic World Bank country office.
ccxxxiv Thomé J-M and Pholsena S (2009) Lao People’s Democratic Republic: health financing reform and challenges in expanding the current social protection schemes. In: Promoting sustainable strategies to improve access to health care in the Asian and Pacific region. Bangkok, Thailand: United Nations Economic and Social Commission for Asia and the Pacific; pp 71-102. http://www.unescap.org/esid/hds/pubs/2529/2529_Chapter%203%20P71-102.pdf
ccxxxv Yap M (2003) On overview of community health insurance initiatives in the Philippines. Manila, the Philippines: World Health Organization; Western Pacific Region. http://www.searo.who.int/LinkFiles/TAG_Meeting_INF5c.doc
ccxxxvi Obermann K, Jowett M, Alcantara M, Banzon E and Bodart C (2006) Social health insurance in a developing country: the case of the Philippines. Social Science & Medicine 62(12), 3177-3185.
109
ccxxxvii Nugroho G and Elliott K (1977) The Dana Sehat programme in Solo, Indonesia. Proceedings of the Royal Society B 199(1134), 145-150. http://rspb.royalsocietypublishing.org/content/199/1134/145.full.pdf
ccxxxviii Saludung D (1997) Community mobilization and empowerment for health. World Health Forum 18(3-4), 274-277. http://whqlibdoc.who.int/whf/1997/vol18-no3-4/WHF_1997_18(3-4)_p274-281.pdf
ccxxxix Thabrany H, Gani A, Pujianto, Mayanda L, Mahlil and Budi B (2004) Social health insurance in Indonesia: current status and the plan for national health insurance [Annex 3]. In: Sein T [Editor] Regional overview of social health insurance in South-East Asia. New Delhi, India: World Health Organization; Regional Office for South-East Asia; pp 101-161. http://www.searo.who.int/LinkFiles/Social_Health_Insurance_an3.pdf
ccxl Suwandono A and Suhardi Q (2002) Case-study Indonesia. In: Brodsky J, Habib J and Hirschfeld M [Editors] Long-term care in developing countries: ten case-studies. Geneva: World Health Organization; pp 119-170. http://www.who.int/chp/knowledge/publications/case_study_indonesia.pdf
ccxli Strauss J, Beegle K, Dwiyanto A, Herawati Y, Pattinasarany D, Satriawan E, Sikoki B, Sukamdi and Witoelar F (2004) Social safety net programmes. In: Strauss J, Beegle K, Dwiyanto A, Herawati Y, Pattinasarany D, Satriawan E, Sikoki B, Sukamdi and Witoelar F [Authors] Indonesian living standards before and after the financial crisis. Singapore: Institute of Southeast Asian Studies – Santa Monica, CA: RAND Corporation, Center for the Study of the Family in Economic Development; pp 316-365. http://www.rand.org/pubs/monographs/2004/RAND_MG137.pdf
ccxlii Sparrow R, Suryahadi A and Widyanti W (2010) Social health insurance for the poor: targeting and impact of Indonesia’s Askeskin program [Working paper]. Jakarta, Indonesia: the SMERU Research Institute. http://www.smeru.or.id/report/workpaper/askeskin/askeskin_eng.pdf
ccxliii Rokx C, Schieber G, Harimurti P, Tandon A and Somanathan A (2019) Health financing in Indonesia: a reform road map. Washington, DC: the World Bank; Human Development Network; Directions in Development. http://siteresources.worldbank.org/INTINDONESIA/Resources/Publication/280016-1235115695188/5847179-1243406029796/Full.Report.en.pdf
ccxliv Barrientos A and Hinojosa-Valencia L (2009) A review of social protection in Latin America. Manchester, UK: University of Manchester; Brooks World Poverty Institute. http://www.ids.ac.uk/download.cfm?objectid=4A21EF26-E9D9-E7FA-9CC79EC2F2EDDC74
ccxlv Mesa-Lago C (2008) Social insurance (pensions and health), labour markets and coverage in Latin America. Social Policy and Development Programme paper number 36. Geneva: United Nations Research Institute for Social Development. http://www.unrisd.org/80256B3C005BCCF9/search/6E08841FF30C8F83C125751200320B25?OpenDocument
ccxlvi Acuña C, Levcovitz E, Vivas G, Rosenberg H, Feo O, Curcio P, Brito P and Urrutia S [Editors] (2004) Exclusion in health in Latin America and the Caribbean. Series Extension of Social Protection in Health No. 1. Washington, DC: Pan American Health Organization; Area of Strategic Health Development; Health Policies and Systems Unit – Swedish International Development cooperation Agency. http://www.lachealthsys.org/index.php?option=com_docman&task=doc_download&gid=222&Itemid
110
ccxlvii International Labour Organization and Pan American Health Organization (1999) Synthesis of case studies of micro insurance and other forms of extending social protection in health in Latin America and the Caribbean [Regional Tripartite Meeting with the Collaboration of PAHO on the Extension of Social Protection in Health to Excluded Groups in Latin America and the Caribbean, Mexico DF, Mexico, November 29 – December 1, 1999 Working document n˚5]. Geneva: International Labour Office.
ccxlviii Bárcena A and Prado A [Editors] (2010) Time for equality: closing gaps, opening trails. Santiago, Chile: Economic Commission for Latin America and the Caribbean. http://www.eclac.cl/publicaciones/xml/1/39711/100604_2010-115-SES-33-3-Time_for_equality_doc_completo.pdf
ccxlix Hujo K and Gaia E (2010) Towards universal social protection. In: Combating poverty and inequality: structural change, social policy and politics. Geneva: United Nations Research Institute for Social Development. http://www.unrisd.org/80256B3C005BCCF9/httpNetITFramePDF?ReadForm&parentunid=36C592406B99BEBAC125778900325E5A&parentdoctype=documentauxiliarypage&netitpath=80256B3C005BCCF9/(httpAuxPages)/36C592406B99BEBAC125778900325E5A/$file/PovRepCh5.pdf
ccl Machinea J, Titelman D and Uthoff A [Editors] (2006) Shaping the future of social protection: access, financing and solidarity. Santiago, Chile: Economic Commission for Latin America and the Caribbean. http://www.eclac.org/publicaciones/xml/0/24080/lcg2294i.pdf
111